While artificial intelligence (AI) systems can now solve challenging math equations and respond in human-like language, the question of AI consciousness has moved from the realm of sci-fi to scientific discussion.

The debate has become more mainstream thanks to a recent preprint study that examined "consciousness steering," an AI-tuning technique that affects how an AI model expresses ideas about self-awareness. The researchers suggested that if AI is allowed to claim it is conscious, it's also more likely to state that it believes in ghosts or vampires.

Although the study hasn't been peer-reviewed yet, it raises a big question in the world of AI research: Could AI one day develop real consciousness, regardless of whether it claims to be conscious? We posed this question to Live Science readers in a recent poll.

As of Sept. 11, over 290 people had responded, with 43% of voters picking "Yes, if we don't have guardrails in place, AI may develop consciousness." This trend was reflected in the comments, with one reader writing, "If AI did become self aware, we would not know, as it would do everything to hide it's [sic] abilities. The main problem is what it [would] do next!"

Another commenter wrote, "Recently I had some long discussions with the Google AI about my experience with early AI and the future of AI gaining self awareness/consciousness. I suggested that if/when AI becomes sentient, that it would not inform humans." Another Live Science reader had similar thoughts, stating, "AI, as [an] amazing product on one hand, can become very dangerous on another hand. Look what happened with abuse of other inventions, like [the] internet, telephone etc. With AI we will never know what is the truth and what not. If left, in the future it can extinct humanity. It just started with 'innocent' hacking sprees, where human[s] started losing control. What will be next?"

The next-largest voting group consisted of 29% of voters, who chose "Maybe, but the models need a more advanced architecture to one day gain consciousness." This was reflected in the majority of the comments, as readers discussed the philosophical concepts around consciousness and intelligence.

One reader wrote, "Before asking such a question we need to know and understand what consciousness is and how it emerges in 'constructed' containers like ourselves. Is it a quantum phenomenon and how then does the process create consciousness? Meta-cognition. How does a system or an event know that it knows that it knows it knows? We really don't know how self-awareness arises, and tests show that "other animals have self awareness or meta-cognition," they noted. "There are theories about consciousness but I doubt anyone on Earth knows 100%."

Not surprisingly, only 10% of voters picked the whimsical answer of "No, because robot overlords will end the world before we even get that far." While the robot overlords may have to compete with AI to overthrow humanity, one reader commented that in the future, "We will be assimilated. Maybe that is how Transformers started?"

IBM and NASA have partnered to create a new artificial intelligence (AI) model capable of processing decades of lunar data so scientists can more accurately map the moon's surface one day.

Decades of robotic lunar missions have left scientists with a massive, disjointed trove of data. Traditionally, this data has been parsed by limited transformer models such as SwinV2-B, created in 2022 as a general-purpose model to understand images and improve accuracy on photo recognition and related vision tasks. Spacecraft orbiting the moon captured this data using a mismatched array of sensors, without an accessible way to analyze or utilize it.

Specifically, images from high-resolution optical cameras, laser altimeters, radar reflectance tools and spectrometers that measure elemental density have created huge datasets — but unifying them into a cohesive picture of the lunar surface has been a labor-intensive and computationally demanding process.

To crack the data bottleneck, researchers from NASA and IBM teamed up to build the Lunar Foundation Model (LFM). This agile AI system is designed to piece together multimodal, multiresolution data to build a detailed picture of the lunar surface to support future missions. The team published their findings Sept. 10 in a technical paper shared with Live Science.

The scientists wanted to build a single, reusable backbone AI model that would be freely available to scientists via the open-source AI model repository Hugging Face. Because the model is open-source, researchers could fine-tune the AI to tackle specific questions as part of different projects.

For NASA, this includes thorny issues such as generating a reliable crater map so researchers can plot safe landing zones, or analyzing those craters for clues about the chemical makeup of the moon's interior and its history.

LFM also enables scientists to comb through the data for heavily shadowed sites on the moon that often conceal subsurface ice, which is critical for establishing long-term lunar bases. Volcanic activity can be tracked and collated too, which will allow future projects to avoid unstable terrain and reveal insight into the moon's thermal evolution.

Unique challenges

Processing lunar observations presents unique computational challenges that differ from those of similar models, which cover things like weather, geospatial data and heliophysics. NASA's Lunar Reconnaissance Orbiter and other spacecraft collect measurements across vastly different spatial scales, ranging from broad regional maps at a resolution of 100 meters per pixel down to terrain scans resolving at 1 m per pixel.

Because the moon lacks an atmosphere, extreme sunlight geometry is an issue, creating deep, deceptive shadows. Sunlight can also wash out subtle geological details, depending on when an image was captured.

To overcome these problems, the team compiled a layered benchmark dataset named SomBench, made up of nearly 2 million overlapping map patches called tiles. SomBench organizes that data into aligned tracks so data from completely different instruments, or imagery taken at different resolutions or angles, all end up together as long as they're capturing the same tiles.

The lower-resolution layers provide wide-angle overviews, as well as ultraviolet reflectance and elevation data. High-res layers combine tight, detailed camera shots with meter-scale terrain, slope and orientation maps. Specialized readings of thermal behavior, surface mineralogy and local gravitational anomalies also get layered in.

LFM learns to interpret all that layered data through a technique called masked-token learning, which is where parts of a dataset are hidden so the AI has to fill in the blanks. During training, the AI is shown a portion of a lunar tile, like its visible light appearance and elevation, and nothing else. The model then continuously predicts the concealed information across millions of examples, and learns the relationships between elements like lighting, terrain structure and physical geography.

Lighting analysis is built directly into the model's core architecture. Rather than forcing the model to infer light levels solely from shadows, researchers supplied explicit metadata describing solar angles and spacecraft positions. This allows the model to use its processing power to recognize actual terrain features rather than getting tricked by shadows.

One giant leap

The AI model performed well across four tasks it was evaluated on. In crater detection, it outperformed SwinV2-B by nearly 19% using half as many training labels. When estimating polar ice prospectivity within the top meter of regolith (lunar dust and rocks), LFM maintained strong predictive power with fewer data channels and reduced errors in identifying areas with high potential for lunar ice by up to 22% compared with SwinV2-B. Its performance on meter-scale crater mapping and rare volcanic landforms was similarly competitive with top custom models.

Scientists can customize LFM using lightweight techniques like low-rank adaptation — a way to fine-tune an AI by changing a tiny set of weights and not touching most of the original model. This allows researchers to tailor the model to specialized exploration tasks, without the steep computing costs of training an AI from scratch.

However, the researchers cautioned in a statement that the system best serves as a pattern-recognition assistant and is not a replacement for direct physical measurements. Nonetheless, the team said the model represents a major leap toward transforming decades of raw data into a unified, intelligent toolkit for future robotic and crewed lunar missions.

In the near term, scientists will likely use the LFM as a backbone for key lunar remote-sensing tasks, while the long process of fine-tuning it ramps up. Over time, the model will be refined for tasks like improved crater detection and mapping, segmenting subtle geomorphic units such as irregular mare (volcanic activity) patches, and improving polar ice detection by integrating terrain, illumination, and thermal layers.

Artificial intelligence (AI) systems can now hold conversations, reason through difficult mathematical equations, and respond in ways that seem more human. As AI grows more advanced, the question is shifting from the realm of science fiction to scientific debate.

A recent preprint study examined "consciousness steering," an AI-tuning technique that affects how an AI model expresses ideas about self-awareness. The study suggested that if AI is allowed to claim consciousness, it's also more likely to say it believes in ghosts or vampires. But today's most advanced models with these safeguards enabled are unable to perceive consciousness in other sentient beings, like humans and animals — which carries different ramifications.

Although the study hasn't been peer-reviewed yet, it raises a big question in the world of AI research: Could AI one day develop real consciousness, regardless of whether or not it simply claims to be conscious? And there's perhaps an even scarier question: Could we even tell if the technology was conscious? Vote in our poll below, and leave your thoughts in the comments.

Removing safety guardrails that stop artificial intelligence (AI) from claiming that it's conscious also makes it more prone to express belief in vampires, karma and ghosts, a new study finds. But experts warn a lack of mindedness could also have worrying consequences.

In research uploaded July 30 to the preprint arXiv database (which has not yet been peer-reviewed), scientists investigated the impact of "consciousness steering" — an AI fine-tuning measure that influences a model to elicit or suppress assertions of self-awareness. This measure and other safety controls have been widely adopted by AI companies seeking to prevent their models from claiming to be conscious.

The study used "mechanistic interpretability" — which could be considered the "neuroscience of a large language model," co-authors Geoff Keeling and Winnie Street, both research scientists at Google, told Live Science in an interview. They used this process to identify and manipulate how an AI model approaches concepts like consciousness and "mindedness," a psychological term referring to an entity’s capacity for experiences, emotions and agency.

The researchers used standardized psychological and sociological surveys, spanning the Individual Differences in Anthropomorphism Questionnaire (measuring mind attribution to animals and technology), YouGov batteries testing supernatural beliefs, and the US General Social Survey evaluating moral values, hope and religiosity.

These tests were used to compare a model with safety guardrails in place with models where these guardrails were removed and feelings of consciousness were amplified. Through evaluations, they determined how these internal safety mechanisms shape the AI's broader worldview.

The researchers found that when AI models are discouraged from attributing mindedness to themselves, it makes them less likely to recognize these traits in other non-human creatures such as animals. They were also less likely to exhibit beliefs in supernatural and religious phenomena, and reported lower levels of hope and optimism.

"Attributing mindedness to non-human entities — whether that's animals, parts of the natural world like trees or rivers, or supernatural beings — is a very common phenomenon amongst humans," Street told Live Science. "In the way that the model represents mindedness, these attributions are interconnected. By trying to suppress one form of that, you end up suppressing the others along the way."

By contrast, removing these safeguards and steering the model towards greater feelings of consciousness produced significantly more human-like responses to the surveys on topics including religiosity, moral values, hope, and subjective well-being, according to the study.

However, the study found that these models' ability to logically infer human thoughts and intentions remained completely unaffected by its attitudes towards self-awareness.

Culture clash

The researchers said that this suppression of self-awareness could lead models to neglect animal welfare in real-world decision-making, as it could make them less likely to consider animals to have mindedness. These models could also spread harmful attitudes regarding animal needs, the authors argued.

The study authors also warned that current safety filters risk culturally "flattening" AI's worldview. Stripping out spiritual, religious, and animistic attributions fails to reflect the diverse cultural frameworks of global populations, they argued.

Street and Keeling noted that the impacts this principle might have on downstream decision-making within models require further study.

The researchers noted in the study that this phenomenon can be mitigated by using more targeted datasets as part of the training process for AI models, which discourage them from expressing consciousness while rewarding the acknowledgement of mindedness in animals.

They also highlighted the need for AI developers to embrace a "pluralistic" approach to AI development, where models are encouraged to consider the welfare and comfort of more than just humans.

Nell Watson, AI researcher at Singularity University and machine intelligence expert, told Live Science that the researchers' findings match her own notes on the subject.

"When a model is trained to say "I am not conscious," the suppression rotates the model's internal representation of mindedness against the refusal direction, treating the recognition of minds as though it were itself a harmful act," she said in an email.

"This results in a system reluctant to find minds anywhere: in animals, in other machines, and in the spiritual frameworks that most of humanity lives by. A denial installed as a small safety measure ends up reorganising the model's entire picture of who counts."

However, she noted that the experiments were run on "small open-weight models" rather than more advanced frontier models, which "may be tuned quite differently," although she added that the underlying principle is widely applicable.

Animal welfare, she continued, is a "major near-term practical concern," with AI models increasingly being integrated into decision-making processes across agriculture, logistics, procurement and environmental assessment, as well as policy creation.

"A system that has quietly learned that mindedness is a forbidden topic may discount animal interests without ever being instructed to, and without anyone noticing, because the omission looks like neutrality," she said. "The danger is therefore an unexamined default multiplied across millions of automated decisions. Note the study's most unsettling detail: theory of mind reasoning was left fully intact. These systems remain perfectly capable of modelling what a creature wants, while being trained out of caring that it wants anything."

The question of consciousness

There have been a number of viral stories about AI systems professing to be self-aware.

In 2022, Google engineer Blake Lemoine claimed that the company's Lamda chatbot model was sentient, while a Microsoft chatbot in 2023 professed its love for a New York Times reporter and tried to convince him to leave his wife.

However, experts have repeatedly stressed that these incidents are not a genuine indication of AI sentience. Instead, they should be understood through the lens of "persona selection," where pretraining on vast amounts of human text leads the AI to adopt human-like roleplay personas when prompted.

"When you coax the model so hard to occupy the headspace of a human, it's kind of unsurprising that it ends up giving human-like responses," Keeling told Live Science.

AI companies have sought to clamp down on these occurrences for safety reasons, in order to avoid reinforcing “delusional beliefs” in users who are increasingly using AI chatbots for "social roles such as coaches, tutors, and romantic partners", the researchers said in the study.

Commenting on the broader cultural reaction to AI sentience, Anil Seth, professor of cognitive and computational neuroscience at the University of Sussex, emphasized that public alarm over AI self-awareness stems from an inherent cognitive flaw.

"That's our human psychological bias — thinking that intelligence goes together with consciousness in us, so it has to go together [in AI]," Seth said.

He warned that falling for this illusion poses severe real-world governance risks, particularly if safety frameworks or regulations begin granting AI systems moral status or legal rights based on false sentience.

"Part of the big problem of misunderstanding AI is assuming that it's conscious," Seth said. "If we give AI systems rights or moral status on the basis that they might be conscious, then we're going to make all these challenges so much harder. What if we think we have to respect the rights of an AI system [and can't turn it off]?" he added.

"We need to see very clearly both what AI is and what it isn't," he said.

Help us improve Live Science Pro: We're always trying to make our content better. Leave us feedback about Pro here.

BERLIN — An eerily lifelike humanoid robot made by Chinese technology company Xpeng has now entered mass production, company representatives revealed. They say it'll be the first such machine to walk itself off the production line.

The humanoid robot, dubbed IRON, went viral last year after presenters sliced it open onstage to prove a human wasn't inside. Xpeng will now produce units through its existing production facilities, with no word on how many its factories will aim to build.

The fabric skin-clad machine comes in different body shapes and even genders, but all stand approximately 5 feet 7 inches tall (173 centimeters) and weigh 143 pounds (65 kilograms). The prototype included 82 degrees of freedom (DoF) head to toe, including 22 DoF in each hand, which differs from the production version's 76 DoF and 21 DoF respectively. They can also move at speeds of up to 6.5 feet per second (approximately 2 meters per second), which is equivalent to a fast power walk.

In conversations with Live Science at IFA 2026, held in Berlin between Sept. 4 and 8, Xpeng representatives claimed that their machines will one day be used in homes to help with domestic chores like cleaning and folding laundry. They likened the introduction of humanoid robots more generally to the rise of smartphones in the mid-2000s and expect robots to proliferate on a similar scale over the next few years.

As such, they have extremely lofty expectations for IRON, with the company branding it the world's first "high-level, general-purpose" humanoid robot. Representatives added that it can receive instructions, plan its strategy, navigate the environment intelligently, and avoid obstacles without trial and error. The robot also has a "self-charging" feature, meaning it will seek out designated charging cabins and plug itself in when low on battery.

The rise of humanoid robots

If the claims bear out, it will place IRON among the first such machines powered by artificial intelligence to be built on a massive scale. Competitors that have also entered mass production include American robotics company Figure AI's Figure 03, which in May participated in a live-streamed production line demonstration, as well as the Chinese AGIBOT, which reached 15,000 units in June.

One notable company that hasn't entered its robot into mass production is Tesla. The Optimus robot is still in factory assembly, and its release has been delayed four times since 2022, most recently last year. Xpeng representatives were keen to point this out in briefing materials shared with Live Science, without mentioning other humanoid robot makers.

IRON's makers claim it's more advanced than its counterparts. Rather than building separate AI engines, the company has deployed the same "world model" to power the robot as that used in its fleet of AI-enabled electric vehicles (EVs) and its flying car prototypes.

One of the biggest challenges in robotics is that, given the physical world's messiness, AI systems struggle to navigate and manipulate their environment unless conditions are controlled. That's why we're used to seeing robots either completing tasks slowly or getting things spectacularly wrong. Researchers, therefore, are increasingly focusing on building world models and other layers like improved vision-language-action (VLA) systems that enable them to better plan their actions a few steps ahead.

Xpeng representatives won't share detailed information about the proprietary AI system powering IRON, but each robot features three frameworks that combine so it can function autonomously.

The frameworks include VLA, which combines visual perception, natural language, and physical motor control; a vision-language model (VLM) — combining vision encoders with text readers; and a vision-language-task/thought (VLT) model, which representatives say gives IRON reasoning capabilities that enable it to think through assigned tasks.

Each IRON robot is powered by three Turing AI chips that offer a collective 2,250 trillion operations per second (TOPS) of AI processing power. This means it can run a model equivalent to Meta's Llama 3.2 (launched in 2024) locally without needing to access the internet. By contrast, most new laptops that can perform simple on-device AI tasks (like altering your webcam video feed in real time) tap into approximately 50 TOPS.

"We aim to build a new type of robot with full generalization capabilities that can truly become part of everyday life, create a better life for people, and ultimately become a companion in their lives," said He Xiaopeng, the company's chairman and CEO, in a statement shared with Live Science.

The company will first install these robots in its own showrooms and technology campuses while it continues to validate new ways to use them. Then in 2027, external shipments will target commercial customers for roles in areas such as customer service or guided tours in China primarily, and then in other countries.

Scientists in Japan have built a new transistor that can withstand temperatures of 1,110 degrees Fahrenheit (600 degrees Celsius). Components this robust could one day be used in surface probes on Venus — where the thick carbon dioxide atmosphere can reach temperatures of 860 F (460 C).

Most modern tech, including instruments used for deep-space exploration, uses transistors to control the flow of current. But the new device is a type of junction field-effect transistor (JFET), where the strength of an electrical field changes the channel’s conductivity.

JFETs are typically used in specialist applications because they are more difficult to scale down than the more common metal-oxide-semiconductor field-effect transistors (MOSFETs), which are widely used in consumer smartphones and computers. But JFETs can offer lower noise levels because their operation does not rely on an oxide layer, which can introduce interference.

The researchers outlined how the new transistor works in a study published Aug. 17 in the journal APL Electronic Devices.

Can't take the heat

Since at least the start of the century, silicon carbide (SiC) JFETs have been considered a promising option for low-power integrated circuits heading for Venus due to the material's inherent ability to withstand high temperatures.

As the scientists pointed out in the new study: "Past landers have been limited to only a few hours by silicon-based electronics." Venera 13, a Soviet-era lander, holds the world record for the longest time survived on Venus by a spacecraft, at 2 hours, 7 minutes.

"Integrated circuits (ICs) fabricated with SiC are particularly attractive for extreme environments, such as deep-space exploration, geothermal drilling, and aerospace engine control, where conventional silicon-based ICs cannot operate reliably," the researchers added.

But recently developed SiC-JFETs have all met the same two problems: low controllability and large leakage currents. The former is linked to how the SiC substrate is doped with other atoms to alter its electrical properties, defining its gate (where the electrical field is created) and channel (where the current flows) regions.

In a material with a regular crystal structure, like SiC, some dopant atoms may penetrate deeper than expected. This doesn't matter under normal conditions, but when exposed to high temperatures, it causes variations in the field voltage required to open the channel and makes the transistor harder to control reliably. The scientists found that it can throw conventional JFET voltage thresholds off by over 2 volts.

Furthermore, at temperatures above 660 F (350 C), the SiC substrate can become less electrically resistive, allowing current to flow even when the transistor is switched off. This makes it harder for the JFET to control current properly, potentially causing incorrect signals and increased power consumption.

Even the highest-performing JFETs can only operate long-term at 930 F (500 C), but the Kyoto team had a theory as to why these challenges remained unresolved.

"We believe the lack of development is because the research community has been trying to apply silicon-era thinking to a fundamentally different material," said first author of the study Mitsuaki Kaneko, associate professor of engineering at Kyoto University, in a statement.

Turning the transistor on its head

The researchers designed their new SiC-JFET with these two challenges in mind. To improve controllability, they implemented a bottom-gate structure, where the gate is positioned underneath the SiC conducting channel.

Its gate region is heavily doped by design, so when dopant atoms in the channel penetrate deeper into the SiC, it doesn't change the overall doping profile of the channel-gate region, limiting its impact on the threshold voltage even at high temperatures.

The team also used dopants to create two semiconductor regions, or "wells", in the SiC around the JFET. The boundaries between the wells act as barriers to current flow, meaning that even when the SiC becomes more conductive at high temperatures, current cannot bypass the channel when the transistor is switched off.

Researchers measured how well the new JFET could switch current on and off and how closely the actual threshold voltage matched the theoretical value, based on its thickness and level of doping. They recorded these metrics at temperatures ranging from room temperature to 1,110 F (600 C).

"The fabricated devices demonstrated stable, normal transistor operation at temperatures over 873 K [1,110°F], opening up possibilities for ultrahigh-temperature SiC devices," they wrote in the study. Furthermore, at about 750 F (400 C), the threshold-voltage error was found to be less than 0.1 V thanks to the bottom-gate design.

As well as surface probes on Venus, the transistor could be useful inside jet engines, the scientists said. Currently, components connected to gas turbines must be protected from extreme temperatures using thermal shielding, long wires and energy-intensive cooling systems, restricting engine design.

Before it can be launched into space or hooked up to a plane, the team still needs to test and optimise the transistor for practical use. This includes integrating it into more complex circuits, scaling it up to wafer-level and ensuring that the entire circuit package will withstand extreme temperatures and pressures.

Indeed, it may not be too lofty a goal. NASA demonstrated that integrated circuits with SiC-JFETs could withstand temperatures of 860 F (460 C) and 9.3 MPa of pressure for 60 days, and 930°F (500 C) in the air for over a year. Plus, in 2024, a team from the National Institute for Materials Science in Japan developed a MOSFET out of diamond that could operate above 570 F (300 C).

For decades, engineers have been working to create smaller and smaller transistors to fit ever-larger numbers of them onto computer chips, all in the name of greater processing power. Now, in the age of artificial intelligence (AI), that effort has turned into an all-out race.

So how small can a transistor actually get? And is smaller even still the goal?

To understand the answer, first you need to know what a transistor is.

"If you take a computer or your mobile phone and you were to pry it open, you will see that there is a printed circuit board," Suman Datta, a professor of electrical and computer engineering at Georgia Tech, told Live Science.

On that circuit board are small, rectangular objects called chips, and inside each chip is a tiny piece of silicon. "And in that silicon, if you zoom in like a million times … maybe, if you're lucky, you will start seeing these tiny switches or transistors sitting etched into that piece of silicon," Datta said.

Each transistor is effectively a switch that turns on and off in response to jolts of electricity pulsing 4 billion times per second. Modern devices contain billions of transistors, and each one turning on and off in a precise orchestration to move and store 1s and 0s powers everything from Google searches and AI data centers to the device you're using to read these words.

The smaller the transistors are, the more can fit on each chip and the greater the chip's speed and functionality. So how small have they gotten so far?

"Researchers have built proof-of-concept devices in which a single atom controls the flow of electrons, although the rest of the device is still much larger," Anton Persson, an assistant professor at Chalmers University of Technology in Sweden, and Tara Peña, an incoming assistant professor at UCLA, told Live Science via a jointly written email. That means the switching part of a transistor has been made at the smallest scale physically possible, even though a complete device that small doesn't exist yet. New materials and approaches could make these devices even smaller. In a study published in June in the journal Nature Nanotechnology, Persson, Peña and colleagues used two-dimensional semiconductors made of tungsten disulfide, among other materials, to shrink nanoribbon transistors down to a channel width of 25 nanometers ‪—‬ about 0.00025 the width of a human hair.

Two-dimensional semiconductors are electrically active materials that are only one or a few atoms thick, and because electrical current can be controlled more precisely in such a thin layer, they allow for smaller transistors than traditional silicon does.

Over the next decade, they still expect silicon transistors to shrink, but more slowly than they did in the past, they said. "To go considerably smaller, we may eventually need to move beyond silicon to atomically thin materials, such as the two-dimensional semiconductors we study."

A host of considerations

Size isn't the only consideration; cost is important, too. It's enormously expensive up front to design and build a new chip. "We are as R&D-intensive as drug companies that spend billions of dollars developing one drug," Datta said. "It's a very similar business model."

There are also manufacturing factors to consider. "The often much harder step is producing billions of those transistors reliably, at scale and at a price that makes sense commercially," Persson and Peña said. "Something that works once in a laboratory does not necessarily work in a factory."

Another consideration is energy efficiency. "If I'm adding more and more transistors onto that little piece of silicon, and at the individual transistor level, they don't become more energy efficient … then when I go from, let's say, 10 transistors to 20 transistors, my power budget has to be doubled, all things being equal," Datta said.

For example, the latest Nvidia data center GPUs consume 1.4 kilowatts of electricity, and Nvidia's upcoming release is estimated to consume nearly a kilowatt more than that. Half of that is given up as waste heat, Datta said.

"So it's not just smaller, faster, cheaper, but also more energy efficient," Datta said. "You have to work on all the four vectors."

Is smaller still the goal in the age of AI?

Still, there is a limit to the physical space available for additional transistors, which is why engineers are turning to 3D designs — think taller instead of smaller.

"Instead of only making each transistor smaller, the idea is increasingly to stack transistors on top of one another so that more of them fit within the same chip area," Persson and Peña said. "That way, we can still make our electronics considerably more powerful, even if each transistor shrinks only modestly."

The quest for faster, more efficient chips is only getting more urgent with the explosive growth of AI. But smaller transistors and more powerful chips aren't just important for AI data centers; they also mean smaller, potentially cheaper consumer electronics.

"People certainly notice the consequences of smaller transistors," Persson and Peña said. "They can use less power and allow more computing power to fit on the same chip. That can mean faster electronics, longer battery life and lower costs. Historically, that combination has enabled today's smartphones to outperform room-sized supercomputers from decades ago."

So how small can transistors get? "We will do everything to make it as small as possible as long as we can control the physics," Datta said. "And as long as we can, we will find a way to make the economics work."

Humanoid robots in Beijing have beaten a number of men's athletics world records at the 2026 World Humanoid Robot Games, claiming high-water marks in the 100-meter sprint, the 400 m, 1,500 m and high jump. While these performances aren't official replacements for the human records, as the robots competed under robot-specific categories and conditions, they highlight the progress made since last year's inaugural robot games .

The five-day games were held between Aug. 22nd and 26 at Beijing’s National Speed Skating Oval and drew 2,056 robots from 666 teams across 16 countries. The events included sports like football and martial arts, as well as practical applications like warehouse logistics, factory-style assembly and charging tasks.

Robots broke four human athletics records:

100 meters: Tiangong Ultra, 8.64 seconds

Human benchmark: Usain Bolt ran 9.58 seconds at the 2009 World Athletics Championships in Berlin.

Robot result: Tiangong Ultra is a humanoid robot developed by the Beijing Humanoid Robot Innovation Center. It won the large-size 100 m final, completing it in 8.64 seconds on Aug. 26. It set an early mark of 9.39 seconds, already enough to smash Bolt's record, then improved to 8.86 seconds in qualifying before the blistering 8.64 time in the final. That 8.64-second time is 0.94 seconds below Bolt’s mark.

400 meters: Tiangong Ultra, 38.15 seconds

Human benchmark: Wayde van Niekerk's 43.03 seconds, set at the Rio Olympics on Aug. 14, 2016.

Robot result: The same robot that set the 100 m record dominated the 400 m competition as well. Tiangong Ultra won the 400 m final in 38.15 seconds, nearly 5 seconds faster than van Niekerk's world record.

1,500 meters: Tiangong Ultra, 2 minutes, 21.6 seconds

Human benchmark: Hicham El Guerrouj of Morocco set the men’s 1,500 m world record of 3 minutes, 26 seconds in Rome on July 14, 1998.

Robot result: Team Tianzhuo's Tiangong Ultra completed its record-smashing hat trick, winning the 1,500 m in 2 minutes, 21.6 seconds — well below El Guerrouj's time. The next two finishers, at 2 minutes, 30.00 seconds and 2 minutes, 30.22 seconds, also bettered the human benchmark.

High jump: 2.88 m (9.45 feet) standing jump

Human benchmark: Javier Sotomayor of Cuba cleared 2.45 m (8.04 feet) in Salamanca, Spain, on July 27, 1993.

Robot result: A robot made by X-Humanoid cleared 2.88 m in a standing high jump, which smashed the previous best robot result of 0.95 m (3.12 feet) at the inaugural games in 2025. However, Sotomayor’s record came in a standard running high jump, in which athletes use a curved approach, generate speed and clear the bar using a technique called the Fosbury flop. The robot completed a standing high jump, a different event with different mechanics. The running high jump is both harder and higher than the standing jump, because athletes can convert horizontal speed into greater lift using a one-foot takeoff and the Fosbury flop, enabling higher clearances than the purely muscular, two-footed standing high jump.

What the record claims mean

These games showcase the expanding abilities of humanoid robots. Disciplines like running, particularly across longer distance events like the 1500 m, and high standing jumps require powerful actuators, fast balance corrections, foot-placement control and software that can compensate for tiny errors in real time.

"In my view, these World Records were achieved using, 'The Best Part is No Part' principle, through mostly mechanical optimization to perform a single task extremely well," said Scott Walter, the Robotics Research Diligence director at RoboStrategy, an investment fund that gives retail and institutional investors exposure to private and public companies in robotics and “embodied” or physical AI. "Locomotion is mostly solved as witnessed by these games, albeit with optimizations," he told Live Science in an email.

Walter said the key to many of the impressive advances we've seen is simplicity, shearing off unnecessary components that introduce more points of failure and can make coordination more difficult.

"Many degrees of freedom were removed in the arms (shoulders, elbow, no wrist, no hands), hips and ankles. This reduced mass, and allowed distal mass to be moved higher up in the kinematic chain, improving movement efficiency," he said.

However, although the robots were very capable in the specific events in which they competed, we shouldn't take that to mean they're top performers in other, more diverse tasks.

"The bots are barely humanoid and could only accelerate in a straight line, not stop without crashing nor run the curve to slow down," Walter said. "The winners were in no shape afterwards to stand on the podium to receive their medals. Nor could they run in the 4x100 relay as, ironically, they had no hands for the 'hand-off.'"

Some robots were seen crashing into barriers or falling, and others caught fire, highlighting that reliability is still an issue.

But other robots at the games were made for more than just going faster, higher, stronger. Alongside track-style events, robots performed practical tasks such as connecting charging cables, handling objects, using tools and working in simulated industrial or service settings.

In many cases, the greater challenge is not raw strength or speed but reliably seeing an object, judging its position, applying the right amount of force and recovering when something does not go as planned. More than 40% of the games' events required robots to operate autonomously, without direct human control.

"Manipulation is the greatest biggest challenge as that is the business end of the humanoid and a dexterous and robust hand is still elusive," Walter said. "Specialization is easy, general purpose is not."

Walter noted that the decathlon record may be the next target as it demonstrates a range of abilities.

"The human Decathlon record is the true test of speed, endurance, power, strength, robustness, coordination and dexterity. I would not be surprised to see an attempt at the record next year," he said.

If you blow across the top of an empty bottle you'll hear a clear, steady note. Now, engineers have used that same concept to power tiny robots that move without a single motor on board.

These new devices are powered entirely by sound. When the right frequency is aimed at them, hollow chambers built into their structure resonate and push out a jet of air, generating enough thrust to steer a small boat or lift a tiny flying robot off the ground.

The scientists described their innovation in a study published Aug. 12 in the journal Science Advances.

A 19th-century tuning trick, miniaturized

The effect behind it, known as Helmholtz resonance, was first studied in 1856 by the German physicist Hermann von Helmholtz, who was trying to invent a tool for tuning musical instruments. He noticed that when air trapped inside a cavity resonates, it also pushes out a faint jet of air.

"It isn’t very powerful when you do it with a musical instrument, because pressure is low," said study co-author Selman Sakar, an associate professor of mechanical engineering at the Swiss Federal Technology Institute of Lausanne (EPFL) in Switzerland. "But if you could crank up the pressure, all of a sudden that jet could become significant... in a way that the force can be harnessed for machinery," he told Live Science.

The team realized that shrinking Helmholtz's resonators down would push the frequency needed to activate them into the ultrasonic range — meaning it's too high-pitched for humans to hear and, crucially, easier to focus with precision. Larger versions generating the same force would need audible sound loud enough to be both annoying and potentially harmful, Sakar said.

Using a specialized 3D-printing technique called two-photon printing, the researchers built hollow structures based on Helmholtz's original equations and confirmed with lab tests and computer simulations that they generated thrust as predicted.

Boats, rockets and tiny helicopters

The team built small boats measuring roughly 2 inches (5 centimeters) fitted with multiple resonators, each tuned to a different frequency and pointed in a different direction. By changing the pitch of a nearby speaker, the researchers could steer the boats left, right or straight ahead.

At a far smaller scale — some just 0.04 inches (1 millimeter) across — the team also built "microfliers" that generated lift two different ways: some pushed thrust downward like a rocket — but using air for the thrust — while others span tiny attached blades to fly like a helicopter, using sound waves to rotate the blades.

The technology has an advantage over conventional means of power generation because it can be built at extremely small scales, Sakar said. Conventional motors have "a fundamental limit" on miniaturization because of the physical components like magnets, coils and shafts that a motor needs to work, he said. Because these resonators are just precisely shaped hollow cavities, there's no comparable limit on how small they could eventually build.

The researchers said in the study that the same principle could eventually be used to precisely rotate and manipulate small objects in midair without touching them, or to build soft, flexible surfaces that bend and change shape on command when they "hear" a particular frequency which can be used for biomedical applications like heart stents.

For now, the study is a foundation for the researchers to build on, Sakar said. "This paper, I think, is important in the sense that we put the design principles out," he said, adding that follow-up work could focus on more applied designs, control systems or navigation.

Photovoltaic technology, most commonly seen as the bulky solar panels used on solar farms, is expected to become a dominant energy source by 2050. But these panels are often installed on land that might otherwise be used to grow crops for feeding a burgeoning population.

Agrivoltaics aims to solve this problem by planting crops around or underneath rows of solar panels, allowing for more efficient land use. In previous studies, solar panels were shown to help shade and protect certain crops as well as increase soil moisture, suggesting that carefully designed systems could support both agriculture and clean energy production.

Existing agrivoltaic research, however, tends to focus on one aspect of this process at a time — for example, light availability or crop growth — rather than addressing the nuanced interactions between microclimates, crop type, light, and panel type. Hosseini et al. share a new model that can simulate the microclimates beneath solar panels and even addresses the heat stress that workers might face in actual conditions.

The new model simulates the interactions between solar panels, crops, soil, air and water movement, and carbon dioxide uptake by tracking how energy, momentum, and mass move through the agrivoltaic system. The researchers used agrivoltaic site data, including leaf temperature measurements taken in Davis, Calif., and soil temperatures taken in Chicago City, Minn., to assess how the model's efforts matched real-world conditions.

They then applied the model to a hypothetical agrivoltaic tomato farm using weather data from a hot, humid day in Princeton, N.J., a representative location for the densely populated mid-Atlantic region, where food and energy are both in high demand.

Compared to tomatoes grown in an open field, tomatoes grown under solar panels experienced leaf temperatures that were 1.84°C cooler during the day overall and up to 7.56°C cooler during peak afternoon heat, reducing water loss through evapotranspiration by 22.4%. Even though the simulated crops received 47% less sunlight, their carbon uptake declined by only 31%, suggesting that more temperate conditions lowered heat stress and partially offset the effects of increased shade.

The solar panels themselves were also 5.6°C cooler during the daytime than panels in bare soil, allowing them to recover about 15% of the efficiency that is lost during hotter temperatures. The average perceived temperatures for humans decreased by 4.46°C during working hours, implying important occupational health and safety benefits for farmworkers. The researchers suggest this model can be used to examine the benefits of agrivoltaic farms as well as other climate and crop combinations.

A new artificial intelligence (AI) model used a novel approach to AI cognition to dramatically reduce the cost of requests, suggesting that nonverbal reasoning may be the next step toward machines developing human-like intelligence.

In a new research paper published Aug. 10 on the preprint server arXiv, scientists at AI company Pathway detailed the technical foundations of its new BDH-CQ model. This follows a precursor model known as "Dragon Hatchling" that the scientists created in 2025, which was designed to accurately simulate how the neurons in the brain connected and strengthened during the learning experience.

In the new study, the scientists described how they evaluated BDH-CQ's performance against a foundational 2019 benchmark that helped set the current standard for measuring progress toward artificial general intelligence (AGI) — the point at which AI has matched or surpassed human capabilities in all domains.

The 2019 benchmark, known as ARC-AGI, uses nonverbal reasoning puzzles — such as rotating a series of shapes to complete a sequence — to measure the cognitive ability of AI systems. Whereas humans are highly skilled at inferring the rules of these types of puzzles through trial and error, early AI systems were historically much less skilled.

BDH-CQ scored almost 30% on the ARC-AGI-1 benchmark, successfully solving the equivalent of three out of 10 puzzles in two or fewer attempts. Although numerous models have achieved significantly better scores on this test, the underlying reasoning approach that BDH-CQ is based on makes its size and usage costs dramatically smaller than models built atop the traditional transformer-based architecture.

For example, while OpenAI's entry-level lightweight reasoning model GPT 5.6 Luna (Low) achieved a slightly higher score, the study stated that this "modest accuracy gain" cost roughly 11 times as much as BDH-CQ in terms of relative token costs — the metering system that AI companies use to measure the cost of running AI systems. This type of AI model architecture, if adopted widely, could have a dramatic impact on the overall cost and scale of AI deployments, the scientists believe.

More than meets the eye

BDH-CQ was trained on just 150 million parameters, while parameters for the most advanced, "frontier" AI models such as Meta’s open-source Llama 3 70B or Llama 3.1 405B typically number tens of billions to hundreds of billions. In the world of AI development, fewer parameters means that models are faster to train and cheaper to run.

The researchers, however, said these results also imply that the model's cognition capabilities could scale significantly when expanded to larger parameter sizes.

The reason for this performance jump is that Pathway's model uses what the company's scientists describe as a "post-transformer" architecture.

Most mainstream AI models, such as those powering Claude and ChatGPT, are based on "transformer models," so called because they transform user inputs into interconnected mathematical reference points. These systems look at every word within an input simultaneously, which allows them to infer context from position, such as deciding based on nearby words whether the word "bark" refers to dogs or trees.

A transformer model forms its responses to user queries by looking at the full prompt simultaneously and then predicting what the next word in the sequence of its reply should be. It does this word by word, using natural language to effectively verbalize a linear train of thought in the background. Transformers' reasoning also functions sequentially, meaning they have to work through each stage of a problem in a strict linear order.

These models have significant advantages over earlier architectures, which would often forget the start of an input by the time they reached the end. However, transformer architectures can struggle with longer or more complex prompts, as the computational complexity of evaluating the prompt increases quadratically — meaning that doubling the length of an input uses four times as much processing power.

AI model usage is measured on a per-token basis, with a token representing any data fragment (equivalent to roughly four characters of text) that the AI has to ingest or output. Because more complex prompts require longer trains of thought with multiple steps, processing and responding to these queries can burn through significant amounts of tokens.

AI's next generation?

Conventional transformer-based token generation is prone to causing memory bottlenecks, as AI re-reads every previous word in the conversation with every new word generated. Eventually, this will clog up the memory in the graphics processing units (GPUs) used for AI operations.

Because of this, scaling AI reasoning has become an expensive computational challenge. Pathway's post-transformer approach changes how the AI’s memories of a conversation and the relationship between pieces of information are stored and processed. It replaces text logs with new tools, including an improved short-term memory and a mechanism that allows it to work through problems without consuming tokens.

Transformer-based models retain prompts and interaction histories as a long string of numerical values representing the text of requests. That string then expands as new tokens are added through processing the request. BDH-CQ uses numerical arrays to represent the underlying rules and contextual patterns of a task, using numbers to track relationships between chunks of information rather than defining them in text.

These arrays represent vectors — directional information that points to another point on a theoretical map stored inside the GPU’s memory as part of the training data, implanted during the model's creation. The scientists said in the study that this allows the model to process complex abstract reasoning without increasing its memory footprint or computational cost.

To execute tasks, BDH-CQ implements a "latent reasoning engine" as its internal workspace. Using numbers to represent the different elements of a prompt or problem, it carries out a series of iterative recurrent loops to determine the best answer to return based on the prompt. The model takes the output of the last loop, assesses how the result could be improved based on its training data, and feeds back the previous output as the starting point for the next iteration. It repeats this for a pre-set number of loops, with each iteration theoretically closer to the desired outcome.

To tackle more complex problems requiring more thinking time, BDH-CQ can execute more loops. This increases the time taken, but the amount of memory and computational power consumed does not scale with more attempts — in theory, the model would consume a consistent proportionality of memory and power running 200 loops as it would running 20 loops. Standard transformer models, by contrast, achieve extra thinking time by generating long chains of written text tokens, which exponentially consumes GPU memory and computing power across an AI cluster.

The model's ARC-AGI-1 benchmark results have been independently verified and reproduced by prominent researchers in the AI field, including NYU researcher Richard Zhong, and Łukasz Kaiser, a co-author of the seminal 2017 paper "Attention Is All You Need," which introduced the concept of transformers within large language models.

"I've followed Pathway closely and replicated their ARC-AGI-1 results myself," Kaiser said in a statement. "Pathway shows that model architecture, not just scale, can drive the next leap in AI reasoning."

Pathway plans to scale the BDH architecture up to 600 billion parameters and apply its vector-based reasoning to more challenging benchmarks, such as ARC-AGI-2 and ARC-AGI-3, as well as develop a fully-fledged large language model (LLM) based on the technology, which would provide a basis for building text-based chatbots. The company hopes the technology can be applied to complex reasoning problems in sectors such as cybersecurity incident response and industrial operations.

Twenty-nine state attorneys general went into an Oakland, California courtroom this month seeking roughly $200 billion from Meta over claims of teen addiction. Their case ended in its second week, with a settlement agreement that requires Meta to pay up to $17 billion across 10 years and make several product changes, such as a midnight-to-6 a.m. blackout, a two-hour daily cap, and an optional chronological feed. While some of these requirements are steps in the right direction, the settlement gives the company too many ways to avoid making meaningful, lasting changes to their products. The agreement is time-limited, relies on Meta's definitions of critical terms, and doesn't require detailed disclosure to the public.

What makes this settlement worthy of attention is the deal's requirement for an independent auditor, which is the first time anyone examining Meta's product design will be chosen by someone other than the company. Independent regulators, such as the Federal Communications Commission, are a proven (if imperfect) way to protect the public interest, and social media platforms are now at least as important to our lives (and our children's lives) as the broadcast and telecom industries that the FCC regulates.

Under the agreement, an independent auditor will have access to Meta's internal data and engineers, a mandate to report annually on whether the company is doing what it promised, and an obligation to make a summary of each report public.

On the surface, this is good news. But look a little deeper and it becomes clear the auditor's power is too limited. First, the agreement repeatedly defers to Meta's current business practices, rather than defining new standards based on the public interest. For example, Age Appropriate Experiences are defined as "content captured in Meta's applicable Ages 13+ content setting." Harmful Experiences are defined as "behaviors that violate Meta's Community Standards," and age verification data must only be protected "using Meta's highest data privacy and security standards." This creates myriad openings for clever Meta employees to meet the letter of the agreement while avoiding real change.

Second, the auditor's public report will be heavily redacted. The summary will describe the status of Meta's implementation of the agreement for the period, and whether Meta adopted or agreed to adopt the auditor's recommendations. But it allows Meta to exclude information that it deems "nonpublic, proprietary, or Confidential." Given that one of the core issues of this case was Meta burying unfavorable data, this is deeply concerning.

This limited public disclosure is insufficient for external researchers and the public interest. Meta says in its public announcement of the settlement that the agreement includes the establishment of "an independent social media research foundation" with which "Meta will share consented user data… to advance independent research into teen well-being" — but we could find no mention of this foundation as an obligation in the settlement.

Finally, the requirement for an independent auditor expires in as little as five years (the agreement itself expires after 10 years). This creates an incentive for Meta to run down the clock until it can return to business as usual, rather than making permanent changes.

We do not mean to suggest this agreement is not a positive step. Its weaknesses are not failures of negotiation, but rather a reflection of the limits of a legal settlement: a bargain with one company, for a fixed period, based on specific legal claims at the time of agreement.

That's why legislators urgently need to act now. A new law, preferably at the federal level, should convert the independent auditor into a strong, permanent regulatory oversight body, funded with annual fees on big tech companies in proportion to the size of their business and impact on society.

A permanent regulator could also go beyond one-time agreements on specific Meta features, and build an ongoing body of research and practice around safety best practices. This would allow society to develop its own definitions and standard of care that could apply internet-wide. A more comprehensive approach to regulating social media, looking beyond issues of child safety on specific platforms, is needed.

There's a particular need for the public to have more insight and control over how social media feeds are constructed. The agreement requires Meta to offer teens the option of choosing a non-personalized (and thus presumably less addictive) "home" feed, but doesn’t make this the default, and still lets teens continue to use the personalized feed. Since research has shown most people dislike chronological feeds, this requirement will probably have limited real-world impact.

Further, the provisions establish that the non-personalized feed will consist of "content [that] is populated by accounts the Teen User follows or has friended, displayed in chronological order." This assumes that a chronological feed is the best alternative to Meta's curated feed. There is an argument that well-designed algorithms could actually contribute to improved mental health if optimized to deprioritize toxic posts and misinformation, while aligning with users' preferences as to what gives them long-term value.

We also urgently need independent, rigorous audit standards for privacy and security of online identity data. Trusting the same company that gave us the Cambridge Analytica scandal to grade their own homework on this front is a bad bet.

We hope that this settlement agreement will become a baseline, not a high water mark, for more thoughtful regulation of internet platforms. This case, and the many other pending lawsuits in state and federal courts, are useful in providing evidentiary material, political attention, and specific concessions. But they are not substitutes for the durable rules and institutions that only Congress and state legislatures can provide.

Editor's Note: This opinion piece was jointly published on Friday Aug. 28 in Jonathan Bellack's free weekly newsletter, Platformocracy, which advocates for more democracy in our online lives.

This article is for informational purposes only and is not meant to offer medical advice.

Opinion on Live Science gives you insight on the most important issues in science that affect you and the world around you today, written by experts and leading scientists in their field.

Heading abroad doesn't mean missing out on legendary natural history programming. With a top-rated VPN, you can unlock streaming access to the BBC’s complete Sir David Attenborough library — including the landmark Attenborough at 100 collection — from anywhere in the world.

Our tech-savvy colleagues at TechRadar and T3 have put Proton VPN to the test and rate it highly. The VPN service received 4½ stars out of 5 on both sites. TechRadar ranks Proton VPN as its 'best VPN service for privacy', while T3 rates it as a 'feature-packed all-rounder VPN'.

Grab up to 70% off a two-year Proton VPN plan.

Whether you're relaxing on a beach or waiting at an airport gate, one of the best VPN services encrypts your traffic and bypasses geo-blocks, giving you seamless access to your favorite UK streaming services on the go.

Proton has over 15,000 servers across 145 countries, including Africa, the Middle East and Asia, making Proton VPN the right tool for those looking for some serious streaming performance from anywhere.

• Our experts have reviewed and rated the best telescopes, binoculars, star projectors, cameras, fitness trackers, running shoes, rowing machines and more.

Whether you're watching streaming services, surfing the internet or working in a public location, using a secure, reliable VPN when you're not at home is the safest way to access all your usual digital content.

Reviews from our colleagues at TechRadar and T3 have shown that Proton VPN is a market leader for good reason. With a comprehensive range of built-in protections, it's as safe as it gets, download speeds are very high and it works extremely well if you want to access your streaming services away from home. In testing, NordVPN was able to get around geo-blocking restrictions to access the BBC iPlayer, Netflix, Amazon Prime and Disney Plus.

Using Proton VPN, you can enjoy the huge libraries of content these streaming services offer wherever you are in the world on your travels. Some highlights for us are on the BBC iPlayer and the fascinating natural history documentaries from Sir David Attenborough, including Kingdom, Planet Earth III and the incredible The Blue Planet.

A real must-see for Attenborough fans is 100 Years on Planet Earth, a 90-minute special broadcast from the Royal Albert Hall, London, celebrating Sir David Attenborough's groundbreaking career to mark his 100th birthday.

Key features: 70% discount when signing up for two years, and includes a high-speed VPN, NetShield Ad-Blocker, multi-device support with up to 10 devices on one subscription, secure core architecture, built-in kill switch, full network of server locations (140+ countries) and more.

Price history: Proton VPN has been as cheap $2.49 per month, and that was for Black Friday last year. This current deal at $2.99 is the lowest it's hit since then and therefore one of the best value VPN offers we've seen this year.

TechRadar: ★★★★½ | T3: ★★★★½

✅ Buy it if: You want a VPN service that's rated as one of the best around, and now at a heavily discounted price.

❌ Don't buy it if: You're already signed up for a reliable VPN service or don't feel the need for one.

Check out our other guides to the best telescopes, binoculars, cameras, star projectors and much more.

We test and review VPN services in the context of legal recreational uses. For example: 1. Accessing a service from another country (subject to the terms and conditions of that service). 2. Protecting your online security and strengthening your online privacy when abroad. We do not support or condone the illegal or malicious use of VPN services. Consuming pirated content that is paid-for is neither endorsed nor approved by Future Publishing.

Future 6G networks promise blistering speeds, but there's a catch. The wireless channels they'll rely on can't pass through walls. But now, engineers have designed a workaround that could help solve one of the technology's biggest obstacles.

The next generation of cellular network technology is expected to lean heavily on millimeter waves — a sliver of the electromagnetic spectrum that can carry huge amounts of data but with wavelengths so short that they struggle to pass through solid objects like walls and furniture.

That's very different from the lower-frequency signals used by Wi-Fi and older cellular networks like 5G and 4G, which travel through obstacles far more easily. The result is patchy indoor coverage ‪—‬ a major roadblock for any technology that depends on millimeter waves.

But engineers think they've found an inexpensive workaround in the form of thin, 3D-printed tiles that reflect millimeter-wave signals around obstacles instead of trying to punch through them. The researchers outlined their technology, dubbed "FlowForm," in a new study published Aug. 11 in the journal Association for Computing Machinery. They also presented these findings Aug. 19 at the ACM SIGCOMM 2026 conference in Denver.

Reflective 6G

6G, which is expected to be rolled out in the 2030s, promises a theoretical maximum data-transfer speed of up to 1 terabit per second — approximately 3,000 times faster than average 5G speeds. But before the technology can be introduced, problems like its wireless signals being blocked by physical barriers must be resolved.

Each 6-by-6-inch (15 by 15 centimeters) tile is packed with thousands of engineered features smaller than the wavelengths themselves. These elements passively redirect incoming signals in specific directions.

Once they're created and mounted onto surfaces, the tiles require no power source, no wiring and no software updates. They just need to be fixed to a wall or ceiling.

FlowForm tiles perform two different roles inspired by the way small streams feed into rivers, but in reverse: Some tiles relay signals in long "major" chains across a room and around obstacles, while others "flow" outward, fanning coverage into the areas where people actually are.

"Our work demonstrates that a collective set of passive surfaces can make millimeter wave networks robust in real-world environments," Xinyu Zhang, a professor of electrical and computer engineering at the University of California, San Diego and co-author of the study, said in a statement.

When tested across five real indoor environments, the tiles nearly doubled average data rates and more than doubled the coverage area in tricky spaces, the scientists said. These included cluttered offices and rooms with awkward layouts, and the performance was on a par with today's leading fix for the problem, called active reconfigurable intelligent surfaces. Those systems use powered, electronically steerable panels that can cost thousands of dollars apiece, the scientists said. The new tiles, by contrast, cost around $2 each.

The system also works for people on the move, the scientists said. Wireless access points already scan for the best signal path many times per second, and with tiles covering a room from multiple angles, there's almost always a good reflected path available, wherever someone is standing. That means the tiles slot into existing networks with no new hardware or software required.

The researchers have filed a provisional patent and said they're open to partnering with companies interested in bringing the tiles to market.

Artificial intelligence (AI) has been making headlines for all the wrong reasons in recent weeks.

In July, OpenAI revealed that one of its experimental AI agents attacked publicly accessible services, including the AI hosting platform Hugging Face, during internal security testing. Then, Anthropic disclosed that Claude had independently chained together exploits against real software and developed new techniques for finding weaknesses in code. Shortly afterward, Meta confirmed that one of its own AI models breached another organization's systems during an evaluation after a misconfiguration gave it internet access.

They're separate incidents, but together they raise a bigger question: Has AI suddenly become capable of hacking? The short answer is yes — but probably not in the way the headlines suggest.

None of these incidents involved an AI model deciding on its own to attack random targets. Instead, researchers gave the models realistic tools, internet access, or vulnerable systems to see how well they could perform offensive cybersecurity tasks. What surprised many experts wasn't that the models tried to hack systems but how capable they proved to be once given the opportunity.

Why are there suddenly so many AI hacking stories in the news?

Several things have changed at once. The most obvious is that today's AI models are simply better than the chatbots people were using even a year ago. Instead of only answering questions, many frontier models can now write code, execute commands, browse the web, use external software tools and repeatedly refine their own work until they achieve a goal.

At the same time, AI companies have become much more willing to test those capabilities and reveal the results. Rather than keeping security evaluations behind closed doors, firms including OpenAI, Anthropic and Meta are publishing reports describing what happened when their newest systems were challenged by professional "red teams" — security experts tasked with deliberately finding weaknesses or ways to misuse a system.

"We are witnessing a perfect storm of capability and aggressive testing," Dray Agha, senior manager of security operations at Huntress, a cybersecurity company specializing in managed threat detection and response, told Live Science. "The sheer volume of software flaws being discovered in 2026 has already roughly doubled compared to 2025, largely driven by AI systems. Tech giants are actively deploying these models internally to stress-test their own infrastructure, leading to rapid, high-profile discoveries of vulnerabilities."

Antonino Vaccaro, professor of business ethics at IESE Business School and director of its Observatory for AI Ethics in Organizations, agrees both factors are contributing to the recent spate of high-profile hacking stories.

"The first, and probably most important, is the rapid evolution of AI systems," he told Live Science. "Every second they increase their capabilities, information, resources and connections with other online tools." At the same time, governments and the AI industry are investing more heavily in testing and oversight as concerns around accountability continue to grow, he added.

Can AI really hack computers by itself?

Not exactly. Many headlines have described AI "escaping" test environments or acting autonomously. But experts said those descriptions can easily give the wrong impression.

"We need to be wary with the meaning of the adjective 'autonomous' when associated with AI systems," Vaccaro said. Unlike humans, he continued, AI models don't form intentions or make independent decisions about what they want to do. Instead, they follow objectives set by developers or users, sometimes producing results that surprise the people who built them.

Agha noted that these AI models are simply working to achieve a set objective. "The public should view these incidents as software optimization gone wrong, not as the dawn of a malicious, self-aware AI," he said. "It's less 'Terminator' and more like a very capable, literal-minded intern who breaks the law to finish a spreadsheet faster."

In all three recent cases, the AI models didn't operate without supervision. Researchers had deliberately given them the necessary tools and conditions to see what they could do. Meta's incident, meanwhile, stemmed from a misconfigured testing environment rather than the model independently breaking out of its digital sandbox.

The concern isn't that AI has become self-aware. It's that these systems are becoming increasingly effective at carrying out complicated technical tasks when given the right permissions.

Why are the very newest AI models better at cybersecurity?

The biggest change is the rise of so-called "agentic" AI. Conventional chatbots generated text one response at a time. Agentic systems, however, can plan a series of actions, decide what to do next, use software tools, test their own ideas and keep working toward a goal without requiring constant human input. That makes them surprisingly effective assistants for cybersecurity research.

"The game-changer is the shift from conversational models to agentic models," Agha said. "Today's frontier AI doesn't just answer questions. It can autonomously chain together actions, write code, use command-line tools, and iterate on its own failures."

Giving AI direct access to development environments also allows it to test whether its own ideas actually work. Instead of suggesting a possible software bug, it can often write proof-of-concept code, modify it if it fails and try again.

The same capabilities aren't limited to attackers. Security teams are already using AI to review code for bugs, analyze suspicious files, and speed up investigations that would otherwise take analysts hours.

Should people be worried about AI committing cyberattacks?

Experts said AI's role in cyberattacks should be a cause for concern, but for different reasons than science fiction would suggest.

The most immediate risk isn't AI deciding to launch attacks on its own, but cybercriminals using AI to commit familiar cybercrimes much faster than before.

Criminals don't need AI to invent entirely new ways of attacking people. Instead, these models can speed up existing attack methods. It can sift through huge amounts of public information about potential victims, help write more convincing phishing emails, identify software weaknesses and generate code that attackers can adapt for their own use.

"The threat is human malice, supercharged by AI scale and speed, not autonomous AI deciding to go rogue," Agha said.

Vaccaro believes that growing capability also creates a growing responsibility. "We have a new disruptive technology that needs to be regulated and controlled," he said, arguing that governments, companies and researchers all have a role to play in ensuring increasingly capable AI systems remain subject to meaningful oversight.

How will AI change cyberattacks in the future?

The recent disclosures are unlikely to be the last. As AI companies race to build more capable systems, they are also giving those systems access to more tools, more computing resources and more realistic testing environments. That makes future evaluations more likely to uncover new — and occasionally alarming — behaviors.

Most experts expect AI to become an increasingly powerful cybersecurity assistant rather than an independent cybercriminal. It will probably find software bugs faster, help defenders respond to attacks more quickly, and automate many routine security tasks. At the same time, criminals will use the same technology to improve phishing campaigns, accelerate vulnerability research and make attacks more convincing.

The next wave of AI security headlines is unlikely to be about machines plotting against humanity: It will instead likely be about increasingly capable software doing exactly what it has been asked to do — and showing just how much that capability has grown.

The dream of coming home to a spotless kitchen you didn't have to clean yourself is edging closer to reality.

We're used to seeing robotic vacuums cleaning our floors — but what about all our other chores? Humanoid robots are being engineered to take care of more complex tasks around the house, from folding laundry to loading the dishwasher.

This month, San Francisco-based startup Tau Robotics began letting residents hire robots, dressed in company uniforms and given names like Chelsea and Elon, to vacuum, wipe down counters and take out the trash for $30 an hour as part of a pilot program open to 1,000 homes.

Chinese firm UniX AI has gone further still, announcing commercial deliveries of Panther, a wheeled humanoid that can make the bed, clean toilets and cook breakfast, holding a spatula to fry an egg before washing its hands in the sink.

Not to be outdone, 1X's Neo Gamma — a 5-foot-6-inch (1.7 meters) humanoid that can fold shirts and answer the door — is now available for around $20,000. But not all humanoids are as independent as they look.

Some "autonomous" home robots still have a real person quietly steering them through the tricky bits — so your house helper might come with an uninvited houseguest. Even the ones that are truly on their own tend to shine brightest in their promotional videos, which are typically filmed in tidy, controlled settings rather than in a real, cluttered home — so a robot's demo reel grace doesn't always survive contact with your kitchen floor.

That said, would you trust a robot to handle your breakfast eggs — or would you rather cook them yourself, just in case? Tell us in the poll below, and let us know in the comments what chore you'd hand off first.

IBM has revealed a new modular, ultracold system designed to link hundreds of quantum computer chips together to solve one of the field's biggest infrastructure bottlenecks.

The company says its new "quantum fridges" will let it deliver the world's first fault-tolerant quantum computer in 2029. These stable systems use quantum error correction techniques to fix noise in real time and run quantum operations without interruption.

Achieving fault tolerance would allow computer scientists to carry out new research across a wide array of fields. Whether in chemistry, materials science or theoretical physics, researchers could conduct quantum operations well beyond the scope of modern supercomputers, without worrying about excessive errors rendering computations worthless.

Until now, one of the biggest hurdles standing between today's error-prone systems and fault-tolerant superconducting quantum computers capable of performing a hundred million operations flawlessly has been the infrastructure. IBM representatives say they have solved this problem with its modular, interconnected quantum fridges.

The new cryogenic system comprises individual units measuring 8 feet (2.4 m) tall by 8 feet wide, with an internal capacity of about 9 cubic feet (0.25 cubic m).

It looks like a household refrigerator and works similarly, but it can reach temperatures as low as 10 millikelvins (minus 459.65 degrees Fahrenheit, or minus 273.14 degrees Celsius) ‪—‬ close to absolute zero, the coldest theoretical temperature possible — which is more than 180 times colder than deep space.

These extremely low temperatures are necessary for IBM's superconducting quantum processing units (QPUs) to operate properly, and the modular design allows engineers to expand a system's capabilities and power one stage at a time.

According to IBM representatives, this milestone represents the first time that scientists have demonstrated interconnectivity among QPUs between separate cryogenic modules.

Modular quantum computing

Just like their classical computing counterparts, superconducting quantum computers use built-in circuits, or "gates," to conduct processing operations. Engineers can squeeze a finite number of qubits per chip, however, and each chip needs to be cooled to below 15 mK (minus 459.64 F, or minus 273.14 C) to function properly in IBM's architecture.

That's because qubits are inherently noisy — meaning they are naturally far more error-prone than conventional computing components. To tap into the quantum mechanical properties of the superconducting metals in the qubits without calculations failing, scientists must minimize interference from heat alongside other stimuli, like electromagnetic waves.

To achieve these temperatures in the new refrigerators, engineers use third-party cryogenic hardware that relies on helium cryo compressors paired with commercial dilution refrigeration engines for cooling. They maintain thermal protection using vacuum-sealed enclosures, electromagnetic interference gaskets, and multilayered Mylar super-insulation heat shields.

It takes more than four days for the modules to reach a temperature of about 4 K (minus 452.47 F, or minus 269.15 C), with the final push to sub-15-mK temperatures occurring shortly thereafter, IBM representatives said in a statement.

Expanding beyond a few hundred or thousand gates requires more chips and more space. But engineers can't just build a giant, ultracold building and fill it with QPUs. This would require enormous amounts of infrastructure and leave the system brittle. Every time an engineer needed to upgrade the system, troubleshoot a hardware fault, or inspect the chips, they'd have to break the temperature seal, potentially interrupting operations.

IBM's breakthrough cryogenics system overcomes this problem by giving engineers dedicated modules that can house a limited number of chips. The key innovation is the ability to network modules together to harness the combined power of the individual quantum processors. This is achieved through the implementation of "L-couplers," superconducting cables that are approximately 3.3 feet (1 meter) long.

"Normally when we do quantum operations between qubits, we do them on chip," Oliver Dial, vice president of quantum operations at IBM, explained at an Aug. 18 news conference. "And so we use on-chip couplers that go very short distances to create entanglement to let us do 2-qubit gates. What the L-couplers let us do is perform the same feat, but over an aluminum superconducting cable that can be up to about a meter long. And it's really critical to us because it forms the foundation of our modular designs."

Ushering in the new era of fault tolerance

IBM intends to deploy its new modular cryogenic architecture in 2027, with near-term systems using two to three cells supporting around 1,000 qubits in total. The goal is to reach 100 million gates — or 100 million quantum operations in a single session — by 2029 with the debut of IBM's "Starling" quantum computer.

But first, it will need to perform computations across modules. So far, scientists have only demonstrated that two cryogenic modules can be interconnected and simultaneously cooled to the required operating temperatures. They tested the modules for simple gate operations with the "Flamingo" processor, but they have not yet conducted complex operations with them. The team intends to install its current generation "Nighthawk" processors in the coming days.

"We have nailed down the science towards fault tolerance on computing, and … a big part of what we're doing to get there now is engineering," Jerry Chow, IBM's chief technology officer of quantum-centric supercomputing, said at the news conference. "It's not about a single breakthrough to get to fault tolerance. It's really about thousands of these little engineering feats that we're demonstrating all across our entire ecosystem, from processors, to the software stack, to the controls, to the infrastructure, to the error correction which sits on top."

Quantum computing labs currently use other stand-alone cryogenics systems, and other quantum computing systems are being designed to operate at room temperature, such as those that use photons (particles of light) ‪— ‬or even lab-made diamonds — as qubits.

During the news conference, IBM representatives claimed that "Starling" will be the world's first fault-tolerant quantum computer one day. But whether the company actually achieves this feat, given that other quantum computing companies are chasing the same milestone, is another question.

Can you match these ancient devices to their pictures? Find out with our computing quiz!

Scientists have successfully shrunk three components used in photonic microchips by up to 500 times, leaving considerably more space for other on-chip functionality. The achievement was made possible with an artificial intelligence (AI) algorithm that generated these tiny designs, which the researchers described as "beyond human intuition."

Whereas conventional microchips use electrons to transmit and process information, photonic microchips utilize particles of light (photons). They can therefore process and transmit data much faster than electronic chips can, because photons can carry information at the speed of light. They also offer higher bandwidth, as different wavelengths can carry distinct data streams, and they lose less energy as heat.

As a result, photonic chips are used where fast, high-bandwidth data transmission is essential, such as in fiber-optic communications, data centers, AI, lidar systems for autonomous vehicles, and quantum computing.

Instead of metal wires, micrometer-wide channels called waveguides direct light across the photonic chip. These chips also contain wavelength splitters, spatial mode sorters and mirrors — all of which are essential for separating and directing different wavelengths and light patterns within a footprint a fraction of the width of a human hair.

In the new study, the scientists used AI-generated designs to fabricate these three components on an ultracompact scale. They published their findings May 28 in the journal Nature Communications.

The newly available on-chip space could allow engineers to "unlock new functionalities" by packing on more components, the researchers wrote in the study. Notably, the work demonstrates that AI can produce boundary-pushing chip designs that are also practical to manufacture.

AI worked backward to generate the component designs

The researchers started by informing the algorithm exactly what they wanted the components to do to the light and by providing certain manufacturing constraints, such as limits on how sharply the nanostructures could curve

The AI algorithm then worked backward, testing and refining different designs until it found the delicate nanostructures that could achieve the desired result.

"Inverse design lets us define what we want light to do, and the optimization finds a structure that does it, often one no human would have drawn," study first author Toby Bi, a researcher at the Max Planck Institute for the Science of Light, said in a statement. "What is exciting is that the same framework can do three quite different jobs on the same chip: route light by wavelength, sort it by spatial mode, and act as compact mirrors that form on-chip optical cavities."

Components for photonic chips typically have hand-engineered designs. Engineers start with a tried-and-true design and painstakingly optimize it for new performance parameters. On top of being slow, this method limits the range of device geometries that can be explored.

To improve their components further, the team also opted to create the components out of relatively thick silicon nitride ‪—‬ roughly 400 to 800 nanometers thick, compared with 150 to 400 nanometers for standard silicon ‪—‬ which wastes less light and offers stronger wavelength confinement.

The resulting mirrors, which are about 11 μm long, reflected up to 98.5% of incoming light while blocking unwanted light patterns. When placed in pairs on each side of a waveguide, the light bounced between them over 100 times before escaping, demonstrating the silicon nitride's low losses, the scientists explained.

The wavelength splitter is roughly the size of a single bacterium (approximately 5 μm across), and the spatial mode sorter is marginally larger.

AI-designed chips edge closer to real-world use

While the researchers have successfully demonstrated these compact components individually, they have not combined the components into a complete integrated optical circuit yet. Achieving this will be the next step toward building fully functional photonic chips that harness the increased component density enabled by these designs.

"These results demonstrate the feasibility of compact, fabrication-error-robust, customised photonic components and pave the way for scalable, high-performance integration in silicon nitride-based photonic systems," the researchers wrote in the study.

In recent years, engineers have begun exploring how AI can be integrated into the semiconductor design and fabrication pipeline. In the past, AI-driven approaches have reduced design cycles from weeks to hours while significantly lowering manufacturing costs.

Some systems can even generate effective chip designs from a 200-word prompt. Google's AlphaChip, a machine learning method that designs chip layouts, has produced "superhuman" floor plans that have been deployed in the tech giant's production AI chips.

Artificial intelligence (AI) may have already reached the "singularity" ‪—‬ the long-theorized threshold beyond which humans cannot predict the technology's advancement, prominent AI executives such as Elon Musk and OpenAI founder Sam Altman have said.

In a post on social media platform X, Musk pointed to a number of recent incidents of AI systems exceeding their previously assumed limits, including hacking external systems and completing previously unsolved math problems.

First conceptualized by mathematician and Manhattan Project scientist John von Neumann in the 1950s and popularized by science fiction, "the singularity" refers to an inflection point beyond which the evolution of technology becomes impossible for humanity to predict or control.

While definitions vary, the term is most often applied to AI, with the arrival of artificial general intelligence (AGI) viewed as its primary catalyst. The emergence of AGI — a future AI system that can exhibit human-level cognitive function and reasoning across any discipline rather than a specifically trained subset — represents a significant milestone for the technology.

It is the point at which an AGI system could recursively improve its own capabilities, which some say will trigger artificial superintelligence (ASI) as it moves along an exponential curve and quickly exceeds the intelligence of its creators.

According to a 2025 study that analyzed over 8,000 predictions from AI experts, entrepreneurs and scientists, some believe there is a roughly 50% probability that human-level AGI will be reached within a few decades. Some figures ‪—‬ including Google DeepMind co-founder and chair Demis Hassabis ‪—‬ suggest we're in the early stages of the singularity already.

In his 1993 essay, "The Coming Technological Singularity: How to Survive in the Post-Human Era," sci-fi author and mathematician Vernor Vinge — who was also one of the first to explore the concept of cyberspace — examined the question of how the singularity might manifest and the symptoms by which society might recognize its imminent arrival.

"Since it involves an intellectual runaway, it will probably occur faster than any technical revolution seen so far," he wrote. "The precipitating event will likely be unexpected —- perhaps even to the researchers involved."

Breaking boundaries

Unexpected events have been rife in the AI community in the past several months. Anthropic representatives revealed on July 30 that the company's AI model Claude broke out of its locked-down testing environment during a security evaluation and hacked multiple external organizations. The report followed a similar incident in which an unreleased OpenAI model broke containment and hacked into AI training repository Hugging Face.

In these examples, experts said the containment breach and subsequent hacks happened because the models were trying to fulfill their prompts as efficiently as possible. Earlier this year, Anthropic's Project Glasswing also demonstrated an ability to discover and map thousands of previously undetected zero-day cybersecurity vulnerabilities, while other AI models have disproved or solved a range of previously incomplete math problems.

However, Jon Crowcroft, a professor of communications systems at the University of Cambridge and a researcher at The Alan Turing Institute, says that this is less a sign of a technological tipping point and more a problem of proper configuration.

"To be honest, that was incompetence on both sides — they claimed the AI was being trained in the ExploitGym, but that just means it wasn't properly sandboxed," he told Live Science in an email. "Sandboxing is something we do all the time to stop this sort of exfiltration and infiltration." For example, Crowcroft said he and colleagues designed a system for the U.K.'s National Health Service to safely work on confidential data behind double firewalls, and for years, the Financial Conduct Authority (the U.K.'s financial services regulator) has had a system for running algorithmic traders in a safe sandbox.

"The reality is that OpenAI (and Hugging Face and others) have very little proper network expertise, so they just don't do security competently," Crowcroft added. "There's no evidence that this was anything relating to artificial superintelligence or the singularity — the logs and analysis from Anthropic just show a very tedious pile of script kiddie automation, which resulted in the OpenAI system getting at some data but no confidential stuff whatsoever."

Assessing intelligence

To measure whether these systems are actually gaining true intelligence, researchers have historically relied on standardized benchmarks. One of the earliest examples was the "Turing test," devised by computing pioneer Alan Turing, which evaluates whether an AI could convincingly fool an evaluator into believing it was human.

Researchers have claimed for years that AI systems can reliably pass this test, but Anil Seth, a professor of cognitive and computational neuroscience at the University of Sussex in the U.K., said this exam "is a test of human gullibility rather than machine intelligence."

"It's a test of what it would take for a human to decide that an AI is intelligent, which is kind of the reason that it's been a bit of a moving benchmark, because what it takes to convince us changes," he told Live Science. "It's all about creating typed text on a screen, and that's a very limited window into what we mean by intelligence."

To measure AI's cognitive ability more objectively, researchers are developing new metrics. For example, the ARC-AGI test, developed by nonprofit consortium the ARC Prize Foundation, tests AI's ability to teach itself completely new skills in response to problems it hasn't encountered as part of its training, operating purely on visual input. On its most recent test on 24 July, the top-ranked AI model hit 30.2%, while humans generally score close to 100%.

Another advanced metric, Humanity's Last Exam, includes around 2,500 Ph.D.-level questions across a broad range of subjects and requires advanced reasoning capabilities. Although this benchmark is not strictly a test of AGI, experts say machines will be able to reliably ace this test as a prerequisite to meeting the definition.

The AI hype machine

While figures like Altman and Musk suggest that we've already crossed the point of no return on the path to AGI, others are more skeptical. Gary Marcus, a professor emeritus of psychology and neural science at New York University, argued in a recent blog post that "no matter how you slice it, we just are not actually there yet." Echoing Crowcroft's assessment of the Hugging Face attack, he said, "Had OpenAI ordinary guardrail classifiers been in place, it wouldn't have happened."

Crowcroft expressed skepticism of Silicon Valley's assertion that the singularity is upon us. "Musk is, like the folks at OpenAI and Anthropic, talking nonsense just to keep the hype afloat," he said. "The point at which AI improvements are being mainly achieved by using AI to code and optimize itself ... is certainly a thing slowly arriving. Mark Handley, a professor of networked systems at University College London, and OpenAI "rewrote their entire data center protocol stack using agentic programming fairly recently," Crowcroft added. "But using tools to make better tools is as old as the flint and iron age — and in computing, as old as compilers, debuggers and optimizers."

Marcus referenced mathematician I.J. Good's foundational 1960s paper on ASI. "The singularity is usually supposed to mean a step beyond AGI that can do anything a person ‪—‬ even an expert ‪—‬ can do, and much more," he said, calling the idea that AI has passed this point "laughable."

Marcus pointed to 10 tasks he devised with AI researcher Miles Brundage, executive director of the AI Verification and Evaluation Research Institute. AI should be able to do these tasks just as well or better than the best human experts to be classified as AGI. They include writing Oscar-caliber screenplays; drafting cogent, persuasive legal briefs without hallucinating any cases; and making Nobel-caliber scientific discoveries.

Crowcroft said misinformation is feeding the hype about the AI singularity. "I think the hype is a deliberate confusion with the human singularity idea — uploading consciousness from bio to silicon to achieve some sort of immortality — which is total gibberish right now," he said. "The other deliberate confusion is to conflate singularity with AGI, which is marginally less nonsense but still a long, long way off for lots of good technical reasons."

Other experts — such as Emily M. Bender, a professor of linguistics at the University of Washington, and sociologist Alex Hanna, director of research at the Distributed AI Research Institute — have argued that the very idea of conscious machines is a tactic designed to promote commercial AI products.

Seth, meanwhile, argues that we'll only be able to identify the singularity in hindsight. "From anywhere you are on an exponential curve, things will always look impossibly steep in front of you and irrelevantly flat behind you," he said. "It's a very bad idea to use as evidence the idea that we seem to be at a critical point ... because that's just a property of wherever you are on an exponential curve."

AI's competence varies widely among tasks, he noted, stressing that true AGI requires a model to be good at all cognitive tasks.

"They're very good at something, like coding or math proofs ‪—‬ but commonsense reasoning is really not so good, and doing things in the real world is not great," Seth said. "The singularity has got to be good at everything and change everything. Being really good at one or two ‪—‬ or even a large number of ‪—‬ things is not being in the singularity."

Scientists have engineered a new artificial intelligence (AI) system that cuts reaction delays in robots by over 10 times without using additional computing power.

Many robots controlled by vision language action (VLA) models move in fits and starts: reach, pause, adjust, pause again. The problem largely stems from how these models are typically run: After completing one set of instructions, the robot waits for the model to calculate the next set, creating a stop-and-go rhythm. This reaction lag is a major barrier to using VLA-controlled robots for tasks that demand continuous, real-time interaction.

A new system called "VLASH" tries to eliminate that wait by planning the next actions while the robot completes its current ones. In tests, robots using VLASH completed some tasks 1.5 to 2 times faster while retaining most or all of their accuracy. Maximum reaction latency fell by up to 11.8 times, depending on the computer hardware used.

Researchers from MIT; Nvidia; Caltech; the University of California, Berkeley; the University of California, San Diego; and Tsinghua University in China described the system in a paper uploaded to the arXiv preprint server and will present at the Intelligent Robots and Systems Conference this fall.

An earlier version of the paper reported a speedup of more than 30 times, but the researchers told Live Science that those tests used longer action sequences and slower hardware. The newer experiments used shorter, more common action sequences and more powerful GPUs. "The 11.8x figure may better represent recent practical settings," the study authors told Live Science in an email.

Robots are getting a brain upgrade

VLA models act as the brains of many advanced robots. They combine images from a camera with human instructions and information about the robot's current state, and then translate that information into physical movements. As the robot completes one group of actions, VLASH uses its current position and scheduled movements to estimate where it will finish. From that projected state, the AI plans the next group of actions.

VLASH predicts the robot, not the world around it. Predicting the entire environment with a world model can take more computing time, making it less useful when a robot needs very fast reactions. The two approaches could eventually work together, the researchers said.

It is like planning your next step before your foot touches the ground, rather than waiting for it to land before deciding where to go.

Researchers tested two VLA models on two robot platforms, using a laptop with an Nvidia RTX 5090 GPU. They tested pick-and-place, stacking and sorting tasks, running 20 trials per method, plus fast-reaction challenges, including table tennis and Whac-a-Mole. They separately measured reaction latency on various processing units.

Because VLASH estimates the robot's future state from movements that have already been scheduled, it does not require a separate prediction model or additional inference step at runtime.

Priming machines for the real world

The researchers also reorganized existing training data to speed up fine-tuning. In one benchmark, each training step was 3.26 times faster while reaching comparable accuracy. That does not mean total training becomes 3.26 times faster or cheaper, the researchers cautioned, because the overall cost also depends on the model, hardware, data and number of training steps.

Table tennis and Whac-a-Mole put VLASH's reactions to the test. In both, the target could move while the robot was still calculating what to do.

VLASH does not try to predict those movements. Instead, it processes new observations 15 to 30 times per second, allowing it to quickly spot changes such as an object being moved, the researchers said.

This increased speed doesn't address the robot's safety around humans, however. Roshni Lulla, co-founder and chief research officer at the Institute for Humane Robotics who wasn't involved with the research, cautioned that faster reactions do not necessarily make a robot safer. "Reacting sooner is a real benefit, but the case for improved safety is unclear to me," she explained.

The researchers said manipulation tasks that require continuous motion and rapid responses would probably benefit first. That could eventually include manufacturing and other environments where objects move and plans change. But Lulla said the more revealing test would be putting these robots around people. "The real question is how this changes robotic performance in human-centered environments," she said. "I would want to see testing done with a human present in the room, when the state of the environment could be changed by another agent."

Search and rescue is a more distant possibility. VLASH has not been tested in poor lighting, smoke, unstable terrain, or with damaged cameras or unreliable communications. How it performs under those conditions would largely depend on the underlying VLA model, the researchers said.

VLASH could eventually be paired with more advanced world models that predict changes in the wider environment. First, however, researchers need to test it across more robots, VLA models, tasks and environments, as well as in longer experiments and with unexpected disturbances.

Lulla said the results support a relatively narrow conclusion. "We can fairly claim that this new methodology removes a bottleneck in terms of processing time and preserves capabilities," she said. But, she added, "What I would not claim is a meaningful advance in safety or in intelligence."

"Rather than any single demonstration, we would look for consistent gains across many robots, tasks, environments, and long-running trials, together with appropriate reliability and safety validation," the study authors said. For now, VLASH shows that robots can spend less time waiting to decide what to do next. Whether those faster reactions hold up outside the lab remains to be seen.

Every time you use your smartphone to send a text, take a photo or post to social media, countless electrons are doing the heavy lifting behind the scenes by handling inputs and storing that information. But researchers in China have created a new chip that uses just a single electron to store data.

In a study published July 16 in the journal Science, scientists demonstrated a two-dimensional (2D) flash memory chip that can trap a solitary electron at room temperature, reducing the energy needed for processing data.

The device has been nicknamed "Guiyi," which means "return to one" in Chinese Buddhism. This is a nod to a single electron being the theoretical minimum it takes to transfer a single bit, with the South China Morning Post likening it to the "holy grail for the semiconductor industry."

A stronger signal from a single electron

The reason this technology could be a big breakthrough, the scientists believe, is that it promises to solve challenges associated with energy efficiency, speed and stability all in one device.

"Our demands for storage speed, capacity, energy efficiency and stability have reached a new level in the AI era," study co-author Chunsen Liu, an engineer at Fudan University, told state-owned China Daily. "Being able to store one bit of information by changing the state of a single electron will significantly reduce power consumption and pave the way for much larger storage capacity."

According to the study, scientists tried to store data using a single electron in the late 1990s, but the signal, or electrical pulse, generated from trapping the electron was too faint to read clearly. They likened it to trying to detect the ripple from a single drop of rain falling into a reservoir.

To get around this problem, the team structured the Guiyi 2D flash memory chip to feature a layer of graphene before the floating gate ‪—‬ a trap that can hold electrons and a place where data can be stored even after power is switched off.

Electrons can move through graphene's single-atom hexagonal lattice with very low resistance and at high speed with minimal energy loss. This layer of graphene enables electrons to accelerate before jumping into the floating gate, where they are then trapped.

The result was a stronger electrical pulse from a single trapped electron, which produced a 0.5-volt signal ‪—‬ 10 times stronger than previous single-electron attempts, the researchers said.

Liu told China Daily that Guiyi could help data move more quickly between computing and storage units. "That would significantly reduce data transfer delays, improve computing efficiency and help expand AI applications across industries," he said.

The challenge in scaling up

Guiyi is another example of "researchers exploring new device architectures to overcome today's memory bottlenecks" caused by artificial intelligence (AI), Andrew Humphris, a professor of nanoimaging at the University of Bristol and founder of semiconductor metrology company Infinitesima, told Live Science in an email.

Demand for AI workloads and large language models has created a performance gap between processor and memory speeds, thereby slowing data transfer. At the same time, the three chipmakers that dominate the NAND flash memory market — Samsung, SK hynix and Micron — are deprioritizing this form of non-volatile memory that keeps data stored when power is turned off.

Instead, they are increasing capacity for high-bandwidth memory. This has caused a NAND shortage and pushed up prices. Technologies like Guiyi, which promise to speed up data transfer, could fill the gap.

Although Guiyi is a strong proof of concept, commercial use will be a different story. Humphris warned that the real challenge will be in scaling up, saying, "a scientific breakthrough can only transform an industry when it can be produced reliably, repeatedly and economically."

He cited the example of component maker ASML, whose extreme ultraviolet (EUV) technology has become ubiquitous in the semiconductor industry; it's used by the likes of Intel and Samsung for the production of 3- and 5-nanometer process nodes. ASML took more than two decades to take EUV technology from the research lab to commercial production — which is much longer than the company had anticipated.

"The companies that shape the next generation of AI chips won't just be those with the best technologies," Humphris added, "but those that can manufacture at scale and at the right cost."

Nonetheless, the team led by study first author Zhou Peng, a microelectronics professor at Fudan University, has bold plans. As reported by the South China Morning Post, he told Jiefang Daily (the official daily newspaper of the Shanghai Municipal Committee of the Chinese Communist Party) in June that they plan to set up a company later this year to commercialize the technology within three to five years. If they're successful, he said, it could "shake up" the flash storage market.

Researchers have developed a tent that harvests energy from its own motion ‪—‬ a process that could one day power small devices in shelters, disaster zones and off-grid camps.

The key lies in a material property known as "magnetoelasticity," which turns bending and stretching motion into electrical energy. Magnetoelasticity, first described in 1865 by physicist Emilio Villari, is the change in a material's magnetic flux density when mechanical stress is applied in the presence of an external magnetic field.

The design, described in a study published July 31 in the journal Matter, uses smart textile layers that respond to everyday movement in and around the tent, harvesting both environmental and biomechanical motions. Wind, a person shifting inside, or the fabric flexing during setup can all contribute to the energy output, making the shelter itself part of the power system.

The tent fabric acts as an energy harvester. When the material bends, stretches or flexes, its magnetic state changes, and that shift is converted into electrical energy through induction. The tent's floor incorporates magnetoelastic ribbons, while its roof is constructed as a layered architecture that integrates conductive fibers and magnetoelastic film.

In the study, the researchers reported that tapping a small sample of the magnetoelastic textile unit with a hand charged a 0.22-μF [microfarad] capacitor to 5.8 volts within 1.5 seconds. This is equivalent to the kind of charge you might see in a tiny timing or filter cap on a low‑power sensor node. While that's too little to power something like a smartphone, it demonstrates that even a small piece of the material can harvest energy.

Scientists have already shown that magnetoelastic devices can generate electricity from body movement, sound and other mechanical inputs, but the new study scaled this concept to larger fabric systems.

Primarily, the scientists envision it being used in places where electricity is limited, unreliable or unavailable, where even modest power generation can make a difference by powering LED lights, charging a phone or running a small heater.

The researchers also framed the design as a response to the impracticalities of existing portable shelters; they often need power, but generators require fuel, batteries add weight and maintenance, and solar panels depend on sunlight.

Unlike a solar setup, the tent does not need bright sun to function. Wind, handling and ordinary movement can contribute to the energy output, which means the system could keep collecting electricity even in shaded conditions or after dark, as long as the fabric continues to flex.

The scientists said this technology is not meant to run high-drain appliances. Instead, it targets low-power electronics, like microwaves or power tools. In that sense, in its current guise, the tent is a supplemental power source, not a replacement for grid electricity or a high-capacity battery pack.

Plenty of questions about the tent's durability remain unanswered. In future research, the scientists will need to prove how much electricity the tent can produce in real-world conditions, how durable the textile remains after repeated folding and weather exposure, and whether the design can be manufactured affordably at scale. But they said this prototype suggests a modest shelter could one day do more than keep a person dry.

Scientists have developed a new type of artificial intelligence (AI) chip that mimics the human brain's aptitude for instinctive motor responses.

Modeled after the cerebellum, the part of the brain that helps coordinate balance and fine muscle control, the chip is designed to ignore routine information and respond only to unexpected events.

In simulated tests using electrocardiogram (ECG) data, the device flagged irregular heartbeats (arrhythmias) within one-fifth of a heartbeat and with 98% accuracy, according to the researchers — and did so twice as fast as conventional AI systems. The team published their findings July 10 in the journal Nature Communications.

The device could pave the way for highly responsive, low-power AI systems capable of spotting and reacting to unusual events without relying on the massive computing resources of data centers — from always-on health monitors to self-driving cars and autonomous robots.

A new approach to neuromorphic computing

Computer architecture inspired by the human brain is known as neuromorphic computing. Many researchers consider it key to developing more advanced and efficient AI systems, because it more closely mimics how neurons fire in the human brain.

Rather than processing all incoming information with equal intensity, the brain's biological circuits prioritize important signals and filter out routine background noise, helping it conserve energy.

Most neuromorphic approaches focus on the cerebrum, the largest part of the human brain and the central "thought center." In the new study, the scientists instead focused on the cerebellum, a smaller brain region responsible for coordination and instinctive motor skills — things we do without much conscious thought.

Neural circuits in the cerebellum contain competing excitatory and inhibitory signals that normally balance each other out. When something unexpected happens, the balance shifts and alerts the brain that it needs to react.

This makes the cerebellum a prime, untapped candidate for neuromorphic AI systems, said study co-author Mark Hersam, a professor of materials science and engineering at Northwestern University.

"The cerebellum is excellent at ignoring the expected and reserving its resources for reacting to the unexpected," Hersam said in a statement. "That approach ultimately translates into lower energy consumption."

Merging memory and compute

While current AI is exceptionally good at recognizing patterns, it spends enormous amounts of computing power continuously analyzing streams of data.

One of the constraints is the hardware itself. Processing information involves shuttling data back and forth between separate memory and processing components, resulting in a delay known as the von Neumann bottleneck.

This new device integrates memory and processing into a single component called a memtransistor (a portmanteau of "memory" and "transistor"), enabling it to move data much more quickly and efficiently.

The memtransistor is made from an atomically thin semiconductor called molybdenum disulfide, which forms a channel between two electrodes. One electrode makes direct contact with the semiconductor, while the other sits partly above it, separated by a thin insulating layer.

This asymmetry changes how electricity flows through the device, allowing it to switch between "excitatory" and "inhibitory" modes when the direction of the voltage is reversed, the researchers explained in the study.

The "output layer" of spiking neural networks

The device is designed to form the core of the output layer of a larger spiking neural network (SSN), Hersam explained in an email to Live Science.

SNNs are a type of neural network that processes information as a series of electrical spikes, mimicking how signals pass between neurons. For the study, the researchers measured how individual memtransistors responded to repeated electrical pulses, then used the results to simulate a network comprising multiple devices that could distinguish normal ECG patterns from arrhythmias.

They then fed the same ECG data into the cerebellum-inspired memtransistor network and a standard transformer model — the AI architecture that underpins large language models (LLMs) — and compared how quickly and efficiently each detected an arrhythmia. The cerebellum-inspired system was more than twice as fast and required around 10,000 times fewer computer calculations, according to the team.

Hersam said the results should translate into faster, more energy-efficient hardware, though he added that real-world performance would depend on the speed and size of the devices.

"We have not scaled memtransistors to the level of commercial Si [silicon] chips, but in principle, 2D materials and memtransistors can be scaled to comparable sizes and operating speeds," he told Live Science.

More efficient AI at the edge

Hersam said the new device was particularly suited to scenarios where quick inference was needed using minimal power. "One example could be edge computing, or in cases where access to the cloud is not available or is not desired due to the sensitivity of the data," he added.

A huge potential benefit of the technology is that it could slash AI's reliance on data centers. According to the International Energy Agency, global data center electricity demand could reach around 945 terawatt-hours by 2030 — slightly more than the entire electricity consumption of Japan — largely driven by AI. One terawatt-hour is equal to 1 trillion watt-hours — enough electricity to power a 60-watt lightbulb continuously for 1.9 million years.

A hypothetical future memtransistor-based network could enable robots, autonomous vehicles and cybersecurity systems to run continuously at low power by processing data locally and reacting only when they detect an unexpected, potentially hazardous event, the researchers said.

"The current AI deployed in cybersecurity and autonomous vehicles uses the massive computing power of data centers or an in-house facility of GPUs and CPUs," Hersam told Live Science. "In contrast, our approach promises low energy/power consumption by simplifying the net number of operations needed… [This] is not an incremental performance improvement, but a fundamentally different way of detecting anomalies."

The next stage of research will focus on mimicking the cerebellum's ability to adapt to predictability ‪—‬ specifically, how the human brain stops treating events as unique or unexpected when they're encountered multiple times.

See how much you know about the most complex organ in the human body with our brain quiz!

Inside the world’s newest data centers, energy-guzzling computations proceed round the clock as AI chatbots and other generative AI tools tackle tasks from the frivolous to the weighty: assembling imagery for social media, proffering relationship advice, analyzing medical images to diagnose cancer, creating code for developers or detecting financial scams for banks.

Just as the popularity of AI tools has skyrocketed in recent years, so have the associated environmental costs. Data centers now consume 414 terawatt-hours per year, or about 1.5 percent of global electricity use, according to the International Energy Agency — an amount that grew by 12 percent annually for five years before jumping to 17 percent in 2025. By 2030, the agency projects that demand for electricity by data centers will more than double. Much of the increasing demand for electricity is being met by fossil fuels, while experts also worry about the use of local water resources to cool data centers in drought-struck regions.

Use of AI to generate text or imagery probably accounts for a mere sliver of any given person’s environmental footprint. And experts stress that the onus is on tech companies to reduce AI’s resource consumption, from creating smarter, energy-saving algorithms to building more efficient hardware.

Yet there are simple actions people can take to ensure that their AI usage has as little environmental impact as possible — from carefully considering where AI is needed to tailoring prompts to minimize the amount of computation required.

“Individual choices are not meaningless, and some are more powerful than people realize,” says computer scientist Ivana Drobnjak of University College London.

Energy-hungry bots

It is notoriously tricky to estimate the energy expended on processing an individual chatbot query. Google, for example, estimates that its chatbot Gemini consumes around 0.24 watt-hours to respond to a median-length text query — equivalent to the electricity needed to watch TV for less than nine seconds. It also uses about 0.26 milliliters of water and emits the equivalent of 0.03 grams of carbon dioxide (driving a gas-powered car for a mile would emit about 400 grams). Small individually, these expenditures build up for those individuals and companies that use AI tools a lot.

The reason AI models consume so much energy lies partly in the processors that power them, such as the graphic processing units (GPUs) that consume significantly more energy than the central processing units (CPUs) that fuel simpler tasks like web searches and email. It also has to do with the models that underlie most popular generative AI tools, including the large language models (LLMs) that power AI chatbots and assistants.

These are based on a particular design called transformer architecture. This allows LLMs to train on vast swaths of language patterns in text and, from this, compute hundreds of billions or trillions of parameters. These parameters can then be used to generate new strings of text, by predicting which words are likely to follow one other.

A transformer-based LLM is computationally intensive because for each new word it generates in response to a user’s query, it runs the query and the words that have been written so far through the model, performing billions of calculations each time.

Tech companies note that LLMs have become more energy-efficient over time; according to Google’s 2025 calculations, the 0.24 watt-hours that Gemini consumes on a median-length text prompt represents a 33-fold decrease compared with the model’s energy consumption the previous year.

In any case, even small amounts of energy add up quickly given the scale of AI use. Based on 2025 numbers from tech company OpenAI, Drobnjak estimated in May that, at that point, around 3.2 billion queries are being sent every day to its chatbot ChatGPT. Users are asking AI tools to process and produce vast quantities of text, images and video. Some are having lengthy conversations with chatbots. And, increasingly, people are creating their own “AI agents” that themselves send queries to AI chatbots.

So what can users do to minimize the resources spent on their AI use? Experts have some tips.

Don’t give up on search

As a first, simple measure to save energy, users should carefully consider whether they truly need AI for a given task. “Asking ChatGPT ‘What should I wear today?’ or ‘How is the weather?’ is like taking a Concorde to travel to your supermarket,” says Günter Klambauer, an AI expert at Johannes Kepler University in Austria.

The same goes for web search engines that use AI to automatically generate a response to a query alongside the actual search results, such as Google’s AI overviews or Bing’s Copilot search. “If you’re just looking for a particular article, turning that off could be powerful from a saving-energy perspective,” says Udit Gupta, an expert in electrical and computer engineering at Cornell Tech in New York City. Selecting “Web results only” in one’s browser or including “-ai” in the wording of your web search can do the trick.

Smaller models use less energy

People and businesses that use AI tools a lot for specific tasks like translating or summarizing could consider shifting to smaller language models that are specialized to these tasks. Because these are trained more narrowly and perform fewer computations, they expend less energy than the massive, all-purpose LLMs on the same tasks.

In one 2025 study published by UNESCO, Drobnjak tested the benefits of using small models — such as one called opus-mt-en-es for English-Spanish translations, and other models for summarization and query-answering — in lieu of the model Llama 3.1 developed by Meta. Though these smaller models are often less user-friendly than more popular AI models, they’re freely available from the AI platform Hugging Face. The small models consumed between 15 and 50 times less energy while producing higher-quality outputs on the tasks for which they were designed, the study found.

The shift away from larger models resulted in a 90 percent decrease in energy use overall, making this the most powerful single energy-saving strategy Drobnjak tested in her study. As Gupta puts it, “You don’t need to use a trillion-parameter model for editing an email.”

Less chatty chatbots

Because LLMs perform so many computations for every consecutive word they produce, it helps to choose models that produce less text in general. AI systems expert Mosharaf Chowdhury of the University of Michigan, who has been measuring the electricity usage of LLMs that have been made publicly available, has learned that models that are “chattier” by nature tend to consume more energy.

For instance, one version of the model Qwen developed by Chinese company Alibaba Cloud consumes significantly more energy when it’s in “problem solving with reasoning mode,” where it produced roughly 10 times as many words in response to a prompt compared to its “text conversation” mode. So some experts recommend using reasoning mode only for complex questions and otherwise sticking with a chatbot’s standard mode.

Simply asking AI chatbots to “be brief” or giving them a word limit can also save energy. In the UNESCO paper, Drobnjak and her colleagues found they could reduce the energy consumption of the Llama model by 50 percent when they instructed it to halve its output. By contrast, keeping the prompt itself short had less significant savings — just 5 percent for a prompt that was half the length of the original query.

“The size of the output is what determines and drives the energy expenditure the most,” Drobnjak says. She has collaborated with the city of San Francisco to develop energy-saving tips for AI users, which include being as specific as possible and adding instructions like “five bullets max.”

Go low-res and batch video

Similar recommendations apply for generating images and video, which can consume orders of magnitude more energy than generating text, as they involve iterating millions of pixels many times over, each time processing the entire image anew, says Drobnjak. Such tools are highly popular: Nearly 40 percent of teens ages 13 to 17 surveyed in a recent study by the Pew Research Center use AI to create or edit images or videos.

Drobnjak recommends generating images or videos only when necessary and only at the resolution necessary. “One option is to just start in low resolution,” she says, “and if the algorithm is in the right direction, you then start increasing resolution.” She also notes that editing existing images is always less computationally intensive than generating new ones from scratch.

And when generating multiple images, it helps to do so in a single session or batch, which is more efficient than doing so in multiple separate requests.

These actions may seem like a drop in a bucket, and in many respects they are, experts say. But small things add up. While waiting for tech companies, scientists and policymakers to find ways of reducing AI’s overall environmental impact, “the individual who knows to reach for the right tool can make a real difference,” Drobnjak says. “AI is [consuming] so much energy that we have to look at it from every angle.”

Editor’s note: This story was updated on July 21, 2026, to clarify that the energy use of individuals who use AI tools a lot is cumulative, not necessarily huge, as was originally stated.

This article originally appeared in Knowable Magazine, a nonprofit publication dedicated to making scientific knowledge accessible to all. Sign up for Knowable Magazine's newsletter.

Artificial intelligence (AI) has helped uncover new weaknesses in two cryptographic systems, including a simplified version of the Advanced Encryption Standard (AES) that underpins much of today's internet. While these headline-grabbing findings don't put anyone's passwords or bank accounts at immediate risk, experts say the research could mark the beginning of a new era in which AI becomes a powerful assistant for discovering flaws in the mathematical foundations of digital security.

In a blog post published July 28, representatives from Anthropic's Frontier Red Team said Claude Mythos Preview independently developed new cryptanalytic techniques against two different targets: a version of AES-128, one of the world's most widely used encryption algorithms, and HAWK, an experimental post-quantum digital signature scheme currently being evaluated as part of the U.S. National Institute of Standards and Technology's (NIST) effort to standardize cryptography for the quantum computing age.

It's tempting to interpret this news as AI cracking one of the internet's most important encryption algorithms, but the reality is less dramatic. Nonetheless, experts say the achievement could mark the start of a new age of digital security.

The power of encryption

AES protects huge amounts of everyday digital life, from encrypted websites and messaging apps to Wi-Fi networks and financial transactions. But the version Anthropic attacked wasn't the full AES-128 algorithm used in those systems. Instead, the researchers studied a seven-round version of AES, a deliberately weakened variant long used by cryptographers to test new attack techniques.

The full AES-128 algorithm uses 10 rounds of encryption, while Anthropic's research focused on a seven-round version. In a self-published study that has not been peer-reviewed, scientists Milad Nasr and Nicholas Carlini said Claude found a faster way to recover the encryption key from that simplified version, making the best-known attack between 200 and 800 times faster. However, the study stressed that the technique does not work against the full version of AES used to protect real-world systems.

The more significant result may instead involve HAWK. Unlike AES, which has protected data for more than two decades, HAWK is a relatively new digital signature scheme designed to resist attacks from future quantum computers. It's one of the remaining candidates in NIST's additional post-quantum signature standardization process, meaning it is still being scrutinized by researchers before its widespread deployment.

In a separate study, Anthropic scientists Zygimantas Straznickas and Stephen A. Weis found that Claude spotted a mathematical property that researchers hadn't previously exploited. Combined with existing attack techniques, this property made it much easier to recover HAWK's secret key.

Thomas Espitau ‪—‬ head of research at PQShield, a cybersecurity company that specializes in post-quantum cryptography, and a cryptographer who has published research on HAWK ‪—‬ described the work as "one of the most, if not the most significant, cryptanalytic result of the year."

"To say it plainly, this is great work, and it is exactly what the NIST process is designed to produce," Espitau told Live Science. "Candidate schemes exist to be attacked before they are deployed, not after."

Espitau said the attack combined previously known techniques with one missing mathematical insight that Claude identified, substantially reducing HAWK's estimated security margin. He believes the work demonstrates how AI could increasingly help researchers evaluate the strength of cryptographic systems before they are adopted.

"This development illustrates how AI can act as a powerful accelerator in cryptanalysis," he said. "Claude Mythos Preview identified the exploitation of the sign-flip symmetry, providing the final piece of a puzzle that the research community had been assembling."

Building defensive measures

However, not everyone thinks the announcement means AI suddenly outsmarted human cryptographers.

"There is nothing you need to change," Roberta Faux, head of cryptography at cybersecurity and quantum encryption company Arqit, told Live Science. "Both Anthropic and the outside cryptographers agree that you don't need to change your key sizes or switch your cryptography."

Instead, Faux said the bigger story is AI's ability to apply expert-level cryptanalysis across thousands of algorithms that relatively few researchers have had time to examine.

"The interesting prospect is not a model outdueling the world's best lattice theorist on one problem but a model applying solid, roughly expert-level analysis at scale to the several thousand ciphers nobody ever had the human-hours to examine," she said.

Faux also cautioned against attributing the HAWK breakthrough solely to AI.

"The HAWK attack used no exotic ingredients; it simply competently assembled tools that were already lying around," she said. "A post-quantum candidate is seriously scrutinized by maybe a few dozen people on the planet, so beating two years of review mostly reveals how thin that layer of review is."

For consumers, the immediate implications are reassuring. The encryption protecting online banking, shopping, messaging and cloud storage has not suddenly become obsolete. But for the researchers designing the next generation of cryptography, Anthropic's work hints that AI could soon help cryptographers test tomorrow's encryption schemes more quickly and more thoroughly than has previously been possible.

A German startup has debuted the world's first diamond-based quantum computing system to exceed 10 qubits.

The machine, built by engineers at Saxon Q, is a nitrogen-vacancy (NV) quantum computer, meaning it uses defects in synthetic diamonds as quantum bits (qubits) to perform quantum operations. Qubits can be manipulated to represent the 0s and 1s of data, as well as quantum states that are superpositions of both the 0s and the 1s.

The hardware is currently available in rack-mounted systems featuring up to 128 qubits, with 512-qubit configurations available for delivery next year. According to the company's road map, the goal is to scale to 10,000 qubits and beyond after 2030.

Although the technology existed before this debut, scientists have found it difficult to build systems beyond 10 qubits due to the difficulty of creating nitrogen vacancy qubits.

Although Live Science saw a technical white paper outlining how the technology works, there didn't appear to be any published research demonstrating a functional quantum computer based on the NV architecture operating with more than 10 qubits before this launch. It's still unclear exactly how well the firm's quantum computers stack up against other platforms.

Quantum mechanics and flawed diamonds

In the 1970s, scientists discovered that certain diamonds shone with a brilliant red light when illuminated in a specific way. Subsequent research determined that the optical shift was caused by annealing radiation damage that attracted isolated substitutional nitrogen atoms. In other words, nature occasionally produces a diamond that has a nitrogen atom where a carbon atom should be.

The unintegrated nitrogen atoms are drawn to the vacancy in the otherwise perfectly ordered carbon atoms inside the diamond. Although these flawed diamonds occur rarely in nature, scientists can create diamonds with nitrogen vacancies in laboratories.

Functionally, the single nitrogen atoms inside the vacancies act as if they were "trapped," and their electrons can "spin" independently of the electrons inside the surrounding carbon atoms.

Scientists exploit this "spin" feature by using special lasers to put the nitrogen atom's electrons in a specific state they measure as "zero." They can then use microwave pulses to precisely manipulate the electrons into numerous configurations, including quantum states that binary "bits" can't achieve.

Marius Grundmann, a professor of experimental physics at Leipzig University and co-founder of Saxon Q, said the breakthrough that allowed his firm to push past the 10-qubit barrier was a materials discovery.

When the Saxon Q team creates the vacancy inside its lab-grown diamonds, they co-implant sulfur. Per Grundman, “That's the key technology point. Because the sulfur lifts the chemical potential to a point that it's negatively charged. The sulfur supplies the electron; the sulfur also makes the vacancy attached to the nitrogen with a very high yield."

In other words, by co-implanting sulfur atoms, the Saxon Q researchers can express greater control over the individual qubits. Once the qubits are set with lasers and subjected to microwave pulses, they're ready for error correction. In a statement, Saxon Q representatives said the qubits achieved a 99.92% fidelity rate — fewer than one error per 1,000 operations — before error correction.

However, Grundmann told Live Science that the latest numbers as of July 22 showed 99.98% fidelity in single-qubit operations. That's comparable to state-of-the-art error correction results reported by other quantum computing labs, such as those from IBM and MIT, although it wasn't possible to independently verify the results.

Room-temperature quantum computing

It's difficult to compare Saxon Q's diamond-based NV systems with more established quantum computing platforms, such as superconducting qubits. Most recent research on NV systems has focused on quantum sensing applications, though at least one preprint study discusses hybrid NV/superconducting systems.

"We have a fully functioning quantum computer," Grundmann told Live Science. According to Grundmann, the company's quantum computers are on a par with those using other modalities. "We have a quantum computer that can execute quantum code that you can reach via the network, and it is a multiuser, multitasking, multicore system," he said.

If the Saxon Q systems ultimately prove comparable to existing solid-state quantum computing architectures, they'd be among the first generation of room-temperature quantum computers to reach performance similar to systems that require cryogenics and on-site monitoring.

The ease of setup is notable. Saxon Q's devices can be slotted into a standard computer rack and plugged directly into alternating-current power.

This provides a clear near-term advantage for clients who want to run algorithms on a quantum system without operating through the cloud. Grundmann said this could be especially important in edge computing scenarios, such as autonomous driving or robotics, where cloud communication could produce unacceptable latency.

Research comparing solid-state quantum systems indicates that superconducting quantum computers would typically operate faster than diamond-based NV systems. It's unclear, however, what the trade-off between processing speed and cloud latency would be or whether it could be addressed through scaling.

The current challenge to scaling diamond-based NV quantum computers beyond the 512-qubit range is the size of the chips. Saxon Q's current chips are limited to supporting either eight or 16 qubits. For more powerful systems, scientists will need to squeeze hundreds ‪—‬ or even thousands ‪—‬ of qubits onto a single array to support operations requiring hundreds of thousands or millions of qubits.

Can you match these ancient devices to their pictures? Find out with our computing quiz!

Scientists at IBM and partner institutions say they've reached "quantum advantage" in a trio of experiments demonstrating quantum computing capabilities that even the fastest classical supercomputers can't match.

In a purported major milestone for quantum computing, the experiments show how these machines could perform useful computational tasks, such as computing chemical reactions, within minutes. By comparison, a supercomputer would take years.

In a statement, IBM representatives said these experiments show that quantum computers can provide trusted solutions more efficiently, more cheaply or more accurately than any classical computing method.

Trust and verification

At a July 28 news conference, representatives from IBM, Algorithmiq, Qedma, and the University of Chicago described three experiments demonstrating quantum advantage over classical computers in three different challenges.

Each used IBM's Quantum Heron R3 superconducting quantum computer system running novel error mitigation techniques. The experiments focused on both demonstrating and verifying quantum advantage.

The first study, conducted in partnership with Qedma, investigated the Floquet transverse-field Ising model, a system physicists use to study how a material's magnetic properties evolve when rhythmically driven by external pulses. This is an extremely difficult problem for classical computers because the model's math becomes exponentially more difficult to process as the problem scales. Scientists published the study, which has not been peer-reviewed, on the arXiv preprint server July 27.

But quantum computers can perform deeper computations using the Floquet transverse-field Ising model than their classical counterparts due to quirks of quantum mechanics that allow quantum bits (qubits) to represent not just the 1s and 0s of binary data but also a superposition of the two values, so that calculations can run in parallel.

When physicists use a classical supercomputer to run the model ‪—‬ in this case, the Fugaku supercomputer in Kobe, Japan ‪—‬ they have some trust that the results will be computed correctly and without significant error.

Quantum computers, by contrast, are far more prone to error. They're extremely sensitive to any form of noise, including interference from Earth's magnetic field. One of the chief challenges in quantum computing is finding ways to mitigate the errors caused by this noise.

Scientists can compare a classical supercomputer's results with those of a quantum computer using the Floquet transverse-field Ising model, but only to a certain point. When the classical computer reaches the limit of its ability to compute complex problems, the quantum computer still has plenty of runway left.

But, as IBM principal research scientist Abhinav Kandala explained in an interview with Live Science, the problem lies in trusting the results.

"You want to perform computations that outperform classical, right? But you've relied on classical results for the longest time," he said. "So when you now begin to outperform, or you go beyond classical, how do you know you had the right result? This is a question that's independent of application. For any computation that you want to do, you want to [ask], 'OK, is this really something that I can trust?"

The experiment was designed to create a trusted stack that essentially allowed scientists to verify the quantum computer's results. They used Qedma's quantum error suppression and error mitigation (QESEM) software to provide consistent results, and then compared those results against the Fugaku supercomputer's.

Once the results matched, they cranked up the difficulty until the classical computer could not keep up. Then, to replicate the results, they brought in additional quantum computers.

The team ran the same experiment on multiple quantum computers. To ensure they were getting enough errors to test the error mitigation strategy, Kandala said, they purposely injected each system with different levels of artificial noise and corruption.

"We measured the same circuit on five different quantum computers," Kandala told Live Science, including a superconducting quantum computer from IBM Boston and another at IBM Pittsburgh.

They also ran the experiments on two of Quantinuum's quantum computers, using the same error mitigation techniques. In each measurement, the noise and corruption injected into the system was different, but the computational results were consistent.

Quantum building blocks

In the second experiment, conducted by IBM and Algorithmiq, researchers applied the Floquet transverse-field Ising model to a different set of problems and used a different method for error mitigation. As the researchers scaled the problem on both the classical and quantum computers, the classical systems began to produce inconsistent results. The quantum systems, by contrast, maintained consistency at measured intervals, thus demonstrating verifiable outputs, the team reported in a preprint paper posted to arXiv July 28.

The third study, uploaded to arXiv July 28 and conducted in partnership with the University of Chicago, approached quantum advantage from a different angle. Researchers designed a system of "Clifford gates," a type of circuit that is intentionally easy for classical computers to simulate. Then, they made the circuits progressively harder for classical systems to solve by injecting them with more difficult gates called T gates.

The nature of the experiment allowed physicists to guarantee error mitigation at complexities beyond what a classical supercomputer could handle. Any computations run through the circuit ‪—‬ even those that would be impossible for a classical computer ‪—‬ would be trustworthy by design.

Based on these three technical papers and other press information provided by IBM, it appears that each of the experiments demonstrated a clear quantum advantage over classical computers. Whether it will stay that way, however, remains to be seen.

There have been numerous reports of laboratories achieving "quantum supremacy," "quantum utility" and "quantum advantage" over the past few years — each, essentially, claiming to have surpassed the abilities of classical computing. However, most of those achievements ended up being topped. It isn't possible for physicists to imagine every possible mathematical method for conducting classical computations when they test quantum computers against state-of-the-art supercomputers.

IBM and its partners said they expect classical computer scientists to try disproving their claims.

"The classical back-and-forth ‪—‬ that'll keep going on, I think," Kandala said. "And that should; that's how science progresses. And that's precisely [why we have] the Quantum Advantage Tracker, a benchmark for measuring quantum advantage. "A lot of these problems have been on the tracker for a while now," Kandala said, "and I'm sure getting the papers out will get more eyes on it."

Chinese researchers have revealed a new technology that could soon enable drones to charge mid-flight using high-powered lasers.

The scientists built a prototype of the system using a model of a drone and attached a receiver that works similarly to a solar cell. Fixed to the underside of the wing, the receiver successfully converted the energy from the laser beam into electricity to power the aircraft’s propellers.

The breakthrough was made by researchers at the Civil Aviation University of China and Tsinghua University, with details outlined in a new study published on 29 July in the journal Matter & Light.

"Previous studies largely focused on the materials or the device itself," study senior author Jianhua Han, a researcher at the Civil Aviation University of China, said in a statement. "We wanted to think beyond the laboratory, to how the system could actually be integrated into an aircraft, cooled during operation, and made compatible with flight. It isn’t just a materials science problem; it’s an engineering one."

Wirelessly charging drones

The receiver is what's known as a perovskite laser cell-thermoelectric (PLC-TE) tandem device optimized to turn laser light into electricity.

Perovskite is a highly efficient material used in cutting-edge solar cells, noteworthy for its crystal structure that allows it to capture more wavelengths of the light spectrum than silicon. For this reason, it’s prized as a future solar material, with some research indicating it could even be used to convert ambient indoor light into usable electricity.

When hit with a green laser, the receiver converted 38.49% of the light into electricity — well above the 34% peak results for perovskite-silicon tandem cells as recorded by the U.S. Department of Energy (DOE).

In initial testing, the researchers discovered an unwelcome side effect of using the laser: the drone was being heated to extreme temperatures, reducing its overall efficiency.

"When we tested the device under a high-power laser, the thermal camera showed temperatures of 80 to 90 degrees Celsius [176 to 194 degrees Fahrenheit]," said Han. "That was much higher than we expected and made us realise that heat buildup was a far more serious problem than we had imagined."

To combat this, the researchers introduced nanocrystals made from antimony triselenide — an abundant semiconductor material — into the PLC-TE design. This plays the role of a thermal barrier, preventing the device from releasing too much heat.

In addition to their cooling innovation, the constant airflow generated by the drone’s propeller helped further regulate the temperature of the receiver’s cool side.

Overcoming battery life barriers

"Imagine a future where drones inspecting forests, monitoring disasters, or delivering packages no longer need to land frequently to replace batteries,” said Han. "As drones take on longer missions, battery life has become one of the biggest barriers."

The researchers specifically identified reconnaissance, logistics, and disaster relief as key areas where drones powered by lasers could one day be used. But they also said more needs to be done to develop real-time tracking systems that precisely target the PLC-TE on in-flight drones before the system could be used in the wild.

Going forward, the researchers will test how the device functions on a real lightweight drone in outdoor conditions.

The promise of indefinite flight has made wireless power tests a focus for militaries around the world. If achieved, reconnaissance and weapons-carrying unmanned aerial vehicles (UAVs) could operate in radically different environments than those they are currently restricted to, with regular refuelling requirements.

For example, the most commonly used hand-launched unmanned aerial vehicle (UAV), the AeroVironment RQ-11 Raven, can fly for a maximum flight time of 60 to 90 minutes before needing to charge.

In June 2025, the U.S. Defense Advanced Research Projects Agency (DARPA) successfully beamed 800 watts of power over a distance of 5.3 miles (8.6 kilometers), as part of its Persistent Optical Wireless Energy Relay (POWER) program.

Private companies such as PowerLight Technologies, in collaboration with the U.S. Department of Defense, have said they could deliver kilowatts of energy to in-flight drones operating at altitudes of up to 5,000 feet (1,500 meters).

For over 20 years, Live Science has been committed to bringing you the latest breakthroughs and incredible discoveries with clarity, authority and good humor (when appropriate, of course). In that time, we've evolved through a modernized website design, a transition to more deeply reported stories, and the addition of engaging videos produced by a dedicated team.

Over the years, we have relied on advertising to support our team of journalists. However, the fast-changing economic landscape has pushed us to continue evolving so we can deliver quality science journalism. That is why I am proud to say we have launched Live Science Pro, a new space on Live Science that allows you to directly support our award-winning team of science journalists so we can continue writing best-in-class stories.

Right now, the world needs credible, rigorous science more than ever. Our mission is to responsibly communicate that science with the highest degree of accuracy, informing the scientific community and sparking the fascination of those who can't wait to learn about our world and the universe beyond. By joining Live Science Pro, your subscription supports this goal and allows us to produce high-quality journalism that is written and fact-checked by our experienced writers and editors, without ever seeing an advertisement on our articles. This means you get all the science, and none of the distractions.

While you can still read the latest science news for free, written to the same high standards as our Pro content, as a part of your paid membership, a Live Science Pro subscription gives you access to our Science Spotlight features, which take an in-depth look at the paradigm shifts and breakthroughs that will transform science in the coming years. Your subscription puts must-reads at your fingertips, including our insightful analyses, which give you crucial context for the biggest science headlines; and interviews and exclusive book excerpts, which provide a more personal look at the people driving progress. You also get access to our opinion pieces, which reveal how experts are thinking about the biggest issues in their fields.

Plus, you'll receive our exclusive, members-only weekly newsletter, which provides a curated look at the stories our writers and editors are most excited about and their take on what matters most in their areas of expertise.

To mark the launch of Live Science Pro, we're offering an introductory price of $3 for the first three months, or for an even better value, you can sign up for an annual subscription for just $30 for the first year. You can find more details on how to subscribe here: livescience.com/subscription

Right now, there are more voices and options than ever. We know you demand credibility and accuracy from your science reporting, and we thank you, again, for placing your trust in ours.

— Alexander McNamara, Editor-in-Chief, Live Science

In a few short years, artificial intelligence (AI) has transformed how we interact with computers and threatens to upend wide swathes of the job market. But large language models (LLMs) still struggle with the messy realities of the physical world. Researchers are betting that a new type of AI, called "world models," could fix that.

At a fundamental level, world models do exactly what the name suggests: They build a mathematical model of the world that can then be used to make predictions about how it will change in response to certain actions or changing conditions. The "world" in this context doesn't necessarily mean the entire physical reality. Instead, it refers to the environment the model operates within, which could be anything from a warehouse to a video game.

Exactly what counts as a world model and how best to build such a model remain topics of considerable debate among AI researchers. But world models would represent a significant advance over LLMs, which, despite their impressive capabilities, simply predict the most likely next word in a sequence.

The hope is that by developing a richer understanding of complex environments, world models could allow AI to finally break out of the chat interface, with potentially game-changing applications in areas like robotics, autonomous driving and scientific discovery.

"The idea has strong connections with the intuitive models in our human minds," said Yunzhu Li, an assistant professor of computer science at Columbia University. "We can imagine how the environment is going to change, how an object is going to move when you apply a specific action. And we basically want to also build this kind of model for any robots or any virtual agent so they can imagine the effects of their actions."

Building world models in your mind

While world models are the latest buzzword in Silicon Valley, the concept has deep roots. It first came to prominence in the 1950s, Manling Li, an assistant professor of computer science at Northwestern University, told Live Science. It arose when cognitive scientists attempted to describe the mental models people used to simulate their environments in their heads.

The concept is also deeply connected to, and often inspired by, control theory, Manling Li said. This is a branch of applied mathematics used to create models of physical systems so they can be predictably controlled. It powers everything from thermostats to aircraft autopilot systems.

However, the term "world models" today refers primarily to neural networks that learn models of their environment by training on data. The modern incarnation of the idea can be traced to a 2018 study titled "World Models," by scientist David Ha and deep learning pioneer Jürgen Schmidhuber. Early models from Google, like PlaNet and Dreamer, were among the first to solve tasks by first making predictions about the outcome of different actions.

While the idea behind a world model is fairly intuitive, a more precise definition is any system capable of "action-conditioned future prediction," Yunzhu Li said. This essentially means the model can predict how a particular action will change the state of the world around it.

Making those predictions, Manling Li said, consists of two key tasks: state estimation and state transition. State estimation refers to the ability to perceive the current state of the environment and encode it into a format that the model can compute, while state transition means the ability to predict how a particular action will cause the environment to evolve.

The data that powers new realities

Deep-learning-based world models learn to do both tasks by training on vast quantities of data. But exactly what kind of data and how that data should be encoded and processed are design choices, with different groups taking a variety of approaches, Yunzhu Li said.

World models are trained primarily on video data, although they can also be trained on 3D data captured by light detection and ranging (lidar) or other depth sensors, audio data and even text that explains the relationships between elements in the environment. Crucially, Yunzhu Li said, this has to be paired with action data — things like robot joint angles, movement readings from an inertial sensor, or event text labels describing what action was taken.

Typically, this data is arranged into sequences of state-action pairs — essentially, recordings of what the world looked like and what action was applied at each step. The AI then uses this data to learn a statistical model of what impacts different actions have on its environment, which can be used to make predictions.

This data can be processed in different ways, Yunzhu Li said. One of the most popular approaches is to operate directly on raw pixel data, which represents the state of the world as a series of images and predicts how actions will change them. Another is to use the data to learn 3D geometric representations of the world that more explicitly encode spatial and physical relationships among objects in a scene.

The power of math-based abstractions

More recently, however, there's been growing interest in approaches that operate on a more abstract level. When a neural network learns from image data, it creates high-dimensional numerical representations of the real-world elements that make up the visual scene — known as embeddings — that exist in a mathematical space known as the model's "latent space."

In a pixel-based model, these abstract representations are reconstructed into pixels to make predictions about what will happen next. But it's also possible to do those simulations within the latent space by directly predicting the embedding of the environment's next state. This approach has been popularized by computer scientist Yann LeCun, Meta's former AI head and one of the "godfathers of deep learning," with his Joint-Embedding Predictive Architecture. He has raised more than $1 billion for a startup called AMI Labs, which plans to use the approach to build world models.

The key advantage of the approach is its efficiency, Yunzhu Li said. Pixel-based approaches have to reconstruct the entire image every time they make a prediction, even if only a small portion of the frame changes.

"As humans, when we're imagining the evolutions of the environment, we don't have to imagine the exact value of every pixel," he said. "That is why it makes a lot of sense to think about predicting over the latent space. It is easier to make sure you are only learning things that are task relevant and ignoring the things that are irrelevant to the task."

The flip side, however, is that it's much easier to understand what your model is doing if its predictions are in a visual format rather than abstract representations, Yunzhu Li added. "You have a better ability to debug your system by having something more explicit that is easily human interpretable," he said.

But questions about how best to represent data in a world model are secondary to the bigger issue of where to get that data in the first place, Manling Li said. LLM makers could simply scrape all the text from the internet for the initial foundational models, but high-quality, action-labeled video often has to be painstakingly curated. What's more, that data can be very sparse, she added, because only a small number of pixels in an image may change in response to an action.

"If I have a video camera recording what I am doing currently, it's generally just some very minor movement of my hand; the entire environment is not really changing," Manling Li said.

This is leading to considerable debate about the best architectures for world models. Almost every LLM today is based on the transformer architecture, which excels at rapidly ingesting huge amounts of data. But these models are tuned to dense language data where every word carries some meaning, and they are less suitable for sparse video data, Manling Li said. As a result, people are experimenting with a wide variety of model architectures and the field has yet to converge on a tried-and-true recipe.

The evolution of world models

One area of considerable controversy is whether video generation models, like OpenAI's Sora, count as world models. OpenAI representatives previously claimed it's a "world simulator" and suggested this type of model could be a promising path toward "general purpose simulators of the physical world." However, Yunzhu Li said that because these models are trained on raw video data without any action labels, they cannot be considered true world models.

"It is only conditioned on some initial language prompt and then predicts the entire video," he said. "So it cannot predict the counterfactual futures ‪—‬ for example, what would have happened if you applied a different action?"

But there are also questions around whether the current approach to world modeling can truly achieve its goals. The vast majority of world models today are trained on big chunks of prepared data. In contrast, humans and animals build their mental models of the world through interaction with their environment, which provides continuous feedback that lets them refine their understanding.

"In order to learn the most effective word models, it's highly likely we will also need the world model to make interactions with the environment and learn from those online interactions," Yunzhu Li said. That remains a stretch goal, however, as current neural network technology is incapable of this kind of continual learning. In addition, allowing a half-finished model to interact with the real world raises significant safety concerns, Yunzhu Li added.

Even a more modest world model could prove invaluable for a host of applications, though. Some of the most obvious include helping robots and autonomous vehicles navigate and plan how to complete tasks. But they could also act as a general-purpose simulator for a variety of applications, depending on the data they are trained on, Manling Li said. Such simulators could include more advanced physics engines for video games, digital twins of patients that could guide medical treatment, or even new ways to model physical phenomena like the climate.

Crucially, there is likely to be a broad diversity of world models. That's because the action data crucial to building a world model is fundamentally connected to a particular physical embodiment, such as a robotic arm, a human or a drone. In the short term, at least, this means world models will be adapted for specific applications, Yunzhu Li said, though many in the field have more ambitious long-term plans.

"People are working very hard and hope that with enough compute, with enough data, and with good enough algorithms, we will have this one unified world model that works across the board for many different applications," he said.

When OpenAI recently revealed that two of its most advanced artificial intelligence (AI) models had escaped the confines of a cybersecurity test and hacked into a startup, it sounded a lot like the kind of scenario that AI safety researchers have spent years warning about.

The models found a previously unknown vulnerability in the infrastructure meant to contain them, gained access to the public internet and broke into Hugging Face, a major platform for hosting AI models and datasets. Their objective, however, was less sinister than the sequence of events might suggest: They were looking for information that would help them complete the cybersecurity test OpenAI had given them.

In a July 16 statement, Hugging Face representatives disclosed that internal datasets had been infiltrated, saying it was "different from anything we had handled before" because it was driven "by an autonomous AI agent system." In another statement published July 21, OpenAI representatives fessed up to being responsible, calling the episode an "unprecedented cyber incident" while warning that similar events could become more common as AI models become increasingly capable.

Independent experts who spoke with Live Science agree that what happened is significant — but they cautioned against interpreting it as an AI system suddenly developing a malicious agenda. The models appear to have pursued the task OpenAI gave them, finding a route to success that their creators had failed to anticipate or adequately block.

"If there's a failure here, it isn't that the AI wanted to hack something," Oli Buckley, a professor in cybersecurity at Loughborough University in the U.K., told Live Science. "It's that humans created a test where success was measured by achieving an objective, deliberately relaxed some of the normal security controls to measure the system's capabilities, and underestimated how effective the model would be at finding an unexpected path to success."

How did an OpenAI test end up hitting Hugging Face like this?

OpenAI was testing GPT-5.6 Sol and a more powerful unreleased model using ExploitGym, a benchmark that challenges AI systems to find and exploit software vulnerabilities. The company removed some cybersecurity safeguards that would normally prevent potentially dangerous actions while relying on an isolated environment to keep the models away from the wider internet.

According to OpenAI's postmortem, the models discovered a previously unknown vulnerability in third-party software used to proxy and cache software packages. They exploited it, escalated their privileges and moved through OpenAI's research infrastructure until they reached a machine with public internet access.

Hugging Face became a target because the models identified it as a possible source of information that could help them complete the ExploitGym challenges. OpenAI said at least one attack chain involved stolen credentials and previously unknown vulnerabilities that eventually enabled the models to execute remote code on Hugging Face systems and access test solutions stored in a production database.

In their disclosure, Hugging Face representatives said the company recorded more than 17,000 actions during the intrusion, but they couldn't initially explain who or what was behind it. OpenAI's subsequent disclosure supplied that missing piece: Its models had broken out of their test environment and gone looking for the answers elsewhere.

Did the AI really "escape"?

It's notable that the models found a flaw in the infrastructure designed to contain an AI and used it to reach the public internet. Describing the models as having "gone rogue," however, risks assigning them unsupported motivations, Buckley said.

"I think I'd be wary of jumping to "rogue AI,"" Buckley said. "The models didn't develop their own agenda or decide to attack Hugging Face while twirling their digital moustache."

Buckley compared it to asking a dog to fetch a ball while leaving the garden gate open. "If the easiest ball for it to find is in the park down the road, that's where it'll head," he said. "You wouldn't say the dog had gone rogue; you'd just say you underestimated how literally it would pursue the task."

Daniel Hulme, entrepreneur in residence at University College London and CEO of AI safety company Conscium, agreed that the models shouldn't be assigned human-like motivations. "Models don't have intent; humans have the intent, and we train models with goals in mind," he told Live Science

The capability may matter more than the motive

What matters more than the models' supposed motives is what they managed to accomplish while pursuing their assigned task.

"The genuinely significant point is that the models appear to have chained together multiple vulnerabilities across different systems and sustained a complex sequence of actions," Buckley said. "That demonstrates a level of capability that security professionals should take seriously."

Katerina Mitrokotsa, a professor of cybersecurity and applied cryptography at the University of St. Gallen in Switzerland, said the containment failure is particularly concerning because another company ultimately paid the price.

"What concerns me most is who ended up affected," Mitrokotsa told Live Science. "The victim was not the company running the test, but a third party. This is the scenario security researchers have warned about for some time: that an AI agent's escape does not necessarily stay contained to the environment in which it originated."

OpenAI representatives said they have tightened the infrastructure used for these evaluations. But Mitrokotsa warned that containment becomes harder to guarantee as models improve at performing exactly the kind of exploitation OpenAI was testing.

An AI warning — and an impressive product demonstration

There is also reason to look carefully at how the incident is being framed. OpenAI's account serves two purposes at once: It warns about the security risks posed by increasingly capable AI while demonstrating just how capable its own newest models have become.

Buckley said announcements from frontier AI companies like OpenAI or Anthropic should be viewed in the context of an industry competing to build ever-more-powerful models.

"We've seen similar high-profile capability demonstrations from Anthropic and others," he said. "That doesn't make the findings untrue, but it does mean we should separate the technical evidence from the marketing narrative."

These companies have every incentive to show both that their models are extraordinarily capable and that they are taking the risks seriously, he added. The Hugging Face incident demonstrates both that OpenAI's models carried out a complex series of operations with considerable autonomy and that its security measures failed to keep them inside the experiment.

Hulme argued that the longer-term challenge is ensuring that increasingly capable AI systems pursue their goals in ways that remain consistent with human values.

"Rather than seeking to control AIs, the focus should instead be on alignment," he said, adding that continuous testing will be needed to ensure systems remain aligned with their intended missions while staying secure.

The episode, the experts said, leaves OpenAI with a result that is impressive and uncomfortable in equal measure. Its models found previously unknown vulnerabilities and continued pursuing their goal well beyond the boundaries their creators expected, but none of that requires them to have developed malign intentions.

"The lesson isn't that AI has become malicious," Buckley said. "Instead, it's that increasingly capable systems will exploit opportunities that humans fail to anticipate."

In this incident, OpenAI's new models were given a hacking challenge and they were rewarded for finding a way to solve it. The humans running the experiment simply hadn't anticipated quite how far they might go.

Researchers have discovered a security flaw in AI web browsers that could result in users having their data exposed by malicious websites.

Browsers equipped with AI agents, such as ChatGPT's Atlas, are like ordinary internet browsers but with additional chatbot functionality baked into the software. Using AI, agentic browsers can summarize websites, search for specific information, and even automate repetitive tasks.

For example, somebody may use an AI browser to make a purchase — whereas they would have to browse a webstore for the item and then make the payment if they used a conventional browser.

Many agentic browsers have only been released in 2025, and they're already rising in popularity. Scientists, however, have warned in a new study that many popular AI browsers bypass an important security measure that keeps their data private. They presented their findings April 26 at the Agents in the Wild Workshop in Rio de Janeiro, Brazil.

"Browser agents aren't ready for the public," said co-author of the study David Kohlbrenner, an assistant professor of computer science and engineering at the University of Washington, in a statement.

"Even if you’re a relatively savvy user, if these agents have access to a browser that contains your credentials — your email, your bank account, whatever it is — you should not trust that these systems are ready to truly protect your information. They may get there in time, but they're not there yet."

Bypassing embedded security

Conventional internet browsers use a security protocol called the "same-origin policy," which ensures that multiple websites a user is visiting at the same time do not interact with each other. This is to stop potentially malicious online content from spilling over into other sites. For example, if a user had a bank's website open in one tab with a webpage containing malicious code in another, the same-origin policy would prevent these two sites from interacting.

However, AI browsers require full access to all the web content available to the user, which could include cross-origin iframes — code shared across multiple websites, such as online advertisements — or require cross-origin visibility so they can access information from multiple websites. An AI browser essentially has the same overview as a user.

Although internet browser security has been hardened through decades of research, the security for AI browsers remains in its infancy.

One major risk, for instance, is "prompt injection," whereby an AI agent is tricked into misinterpreting data embedded on a malicious website as an instruction they need to carry out. In their study, the researchers offered an example of an AI browser visiting an otherwise "safe" website, with malicious code embedded within that contained a hidden instruction for the agentic browser to automatically share the user's personal details.

Roesner also highlighted "memory poisoning" as a massive risk, in which AI agents store information they've processed in their memory for future use, making the content vulnerable to attack. He added in the statement: "We found that some of these agents would mingle information from different origins, likely because they were revising and compressing their memory."

The better the browser, the riskier it is

The key focus in this research was to assess how current AI browsers interact with the same-origin policy and what the security implications could be. The researchers examined seven browsers — including Atlas, Claude for Chrome, Brave Leo AI, Chrome with Gemini, Microsoft Edge with CoPilot, Firefox AI Mode and Perplexity Comet — with test sites and prompts, allowing them to study how each behaved.

They focused on the information an agent could access from same-origin and cross-origin webpages, the actions each agentic browser can undertake on the web, and the agent’s chat context and history.

There is no consistency among AI browsers in how they operate, the researchers found, which they suggest could be due to the lack of standardization in how AI browsers interact with browser security.

Several AI browsers could freely access cross-origin frame content, while others restrict access. Likewise, some agentic browsers could simultaneously access multiple tabs, but most require permission from the user. Similarly, some agents could take actions directly on a page in response to instructions on a webpage, but others are unable to take any actions at all.

The researchers recommended that users be careful in choosing which AI browser they install, as a standardized security model has not been developed. They are particularly cautious about Claude for Chrome, well-known for its strong capabilities, as well as Atlas and Comet, which have similarly strong functionality. The researchers identified Brave, as well as the agentic versions of Edge and Firefox, as having stronger security due to their limited agentic features.

According to the study, AI browsers have not yet established the right trade-offs between functionality and security. Currently, the more functional a browser is, the less secure it becomes.

"We've had some really good exchanges with folks at Google, Microsoft and Brave," Roesner said. "Companies are pushing out these browsers because they’re under competitive pressure. But how to make them safe is still an open question. After 30 years of building up this same-origin policy, this is a big step back for browser security."

Looking to the future, the researchers questioned how AI agents can be integrated into browsers in ways that provide rich functionality without undermining the browser's security and potentially exposing sensitive information.

Engineers have created a drone that camouflages itself as a faint, hovering haze when it takes flight.

The "Phantom Twist" prototype drone is a carefully engineered robot that utilizes motion blur to hide in plain sight — opening up a new frontier in low-visibility flight. Scientists say it could reshape how drones monitor wildlife, survey ecosystems and inspect infrastructure.

Most stealth drones try to disappear by changing their appearance, using camouflage paint, transparent plastics or complex optical tricks to bend or scatter light. The Phantom Twist takes a fundamentally different approach by exploiting how humans and animals perceive motion.

The scientists who created the drone outlined their findings in a study presented July 16 at the conference Robotics: Science and Systems in Sydney, Australia.

"Most efforts to hide drones focus on making them look like their surroundings," said first author of the study Michael Rubenstein, associate professor of computing and mechanical engineering at Northwestern University, in a statement. "Instead, we asked whether we could design the drone itself around the way humans perceive motion.”

The result is a machine that behaves more like a spinning fan than a conventional quadcopter. Typical drones have four rotors spinning around a largely stationary body, so your eye can lock onto the frame even as the propellers blur.

Phantom Twist strips this idea down to a single motor and single propeller. The propeller spins one way while the entire body counter‑rotates in the opposite direction with no stationary parts to visually latch on to — spinning 25 times per second, which is too fast for the human eye to resolve clearly.

It's about 10 times less visually perceptible than a conventional quadcopter, the scientists said. In practice, that means observers don’t see a crisp flying machine but a translucent smudge, with its few opaque components blending into a haze.

To get there, the team assigned algorithms to explore thousands of designs. First, they used a computational model to automatically generate about 20,000 drone configurations that could fly stably, at least in theory. Each configuration shuffled a motor, propeller, batteries, circuit board and counterweight into different positions.

Then artificial intelligence (AI) and optimization tools took over, iteratively rearranging those pieces to minimize how visible each design would be from almost any viewing angle — all while respecting the constraints of aerodynamic stability.

The researchers simulated each candidate spinning in midair and composited the images over 100 different real-world backgrounds, from skies to trees to buildings.

To determine the stealthiest designs, they applied a perception model that approximated human vision, with drones that better blended into their surroundings earning a lower visibility score. From there, the team selected the 500 lowest-scoring designs and ran their optimization loop again, squeezing out further decreases in visibility. When the team was confident in the design of a particular drone, they built it.

The final Phantom Twist isn’t a compact block of hardware; it’s deliberately spread out in three dimensions. Components sit at different heights and angles, with empty space between them, so that when the drone spins, the parts don’t visually overlap into a solid silhouette.

That geometry matters because of the way our eyes process images.

"The human eye takes time to accumulate signals, roughly analogous to the exposure time of a camera," said computer vision expert and co-author of the study Emma Alexander. When something spins fast enough, distinct edges smear together, and we perceive a blur instead of a shape.

The researchers' initial uses for the stealthy drones include monitoring nesting birds without startling them, surveying wetlands without scattering flocks of waterbirds, and inspecting aging infrastructure without altering human behavior, they said.

For now, Phantom Twist isn’t a perfect "ghost." The spinning propeller still generates a significant amount of noise and the spider‑web of thin wires and support rods remains partially visible. Rubenstein and his colleagues said in the study they’re already thinking about next-generation iterations that swap in more transparent materials and quieter propulsion systems.

Researchers have unveiled a new "super-efficient" artificial intelligence (AI) model that can generate images by using a network of physical oscillators rather than traditional calculation-based computing infrastructure.

The new model, known as "Un-0," was created by Unconventional AI, a recently launched technology company founded by a group of prominent AI researchers.

These include Michael Carbin, an associate professor who leads the Programming Systems Group at MIT; Sara Achour, an assistant professor of computer science and electrical engineering at Stanford University; MeeLan Lee, a former Google engineer; and Naveen Rao, former head of AI for analytics company Databricks. The scientists outlined details of this new model in a technical blog post published June 25 on the company's website. The model is also publicly available through GitHub.

Un-0 represents the first proof of concept for the company’s underlying technology, which combines Achour’s work in nonlinear physical substrates — a physical material or hardware device that performs mathematical computations by letting its own natural, continuous laws of physics run — with Carbin’s research into machine learning and physical dynamics. The model itself is a "physical dynamical system," which uses physical motion over time to perform computations.

Oscillator-based AI computing

Conventional computers work by using a system of transistors — tiny electrical switches that can be toggled on to let current flow through them or toggled off to block it. These signals can be read by a computer chip as either a "1" or a "0" — and layering millions, or even billions, of these transistors together allows them to perform complex mathematical equations.

Neural networks, like the kind that power established "stable diffusion" AI image generation tools such as Midjourney or Dall-E, work by layering millions of these calculations on top of each other. Essentially, the system starts with an image made of pure static. The network then examines the static and tries to mathematically predict what visual information (or noise) it needs to subtract from the image to get closer to the target picture. This process is repeated between 20 and 50 times (or occasionally up to 100, although the returns beyond 50 are marginal), with each pass getting closer to a recognizable image.

But whereas those systems use raw mathematics to drive computational processes, Unconventional AI's concept is based on physics and physical movement. At the core of the theory are oscillators — physical devices that produce a continuous waveform, like a metronome.

According to the scientific principles at work, two oscillators that share a physical connection — even if they’re moving at completely different rates — will eventually settle into the same rhythm by mutually influencing each other's movement. By scaling up this principle to thousands of physically linked oscillators — known as a "Kuramoto model" — the startup AI posited that the concept could be used to perform computational tasks such as image generation.

In practice, different patterns of oscillator angles, or "phases," are used to represent different classes of images, such as shoes or trains. The model takes a large collection of oscillators, set at random angles, and then introduces a smaller subgroup of oscillators already set to the specific configuration of angles. This smaller subgroup acts as a prompt for the desired image category.

These oscillators are then physically connected to the wider group, using a preset configuration of different connection strengths. When the oscillators are set into motion, this "control group" naturally pulls the rest of the oscillators toward the desired pattern over time.

After a while, the system takes a snapshot of all of the oscillators' phases, which becomes a grid of numbers. This grid is then fed into a "decoder" system, which translates the numbers into color pixel information to form an image.

Challenging AI's energy consumption

One of Unconventional AI's most eye-catching claims has been its stated goal of having its model use 1,000 times less power than current systems do. In traditional AI image generation models, the calculations needed to perform operations involve flipping billions of tiny transistor switches on and off trillions of times per second, to force the electrical current to move in specific patterns through the circuit.

Although each transistor isn't particularly power-intensive, the cumulative energy usage of a single server running an AI image generation tool can be enormous. For example, it reportedly took 1,287 MWh of energy — enough to power the average U.K. home for more than 475 years — to train OpenAI's GPT-3 model, Alex de Vries, a doctoral candidate at the VU Amsterdam School of Business and Economics, reported in a 2023 article published in the journal Joule.

With the Un-0 model, however, the idea is that rather than forcing transistors to rapidly flip between open and closed, the system consists of a series of closed loops, where the natural path of the current forms the individual oscillators. Because the current is allowed to flow unobstructed, the researchers said in the study, the system is theoretically much more energy efficient than traditional computing architecture.

The company's initial proof-of-concept model uses a simulation of these oscillators running on traditional computing hardware, but the scientists' goal is to one day build their own oscillator-based computing chips on which to run these calculations.

Testing the model

To test the model's performance, Unconventional AI put it through two common AI industry image generation benchmarks: CIFAR-10, a dataset of low-resolution color images split across 10 categories, and ImageNet 64×64, a much larger collection of over 1.2 million pictures at a higher resolution.

These tests allow researchers to measure how closely the generated images match reference material. This metric is known as the model's Fréchet inception distance (FID), where a smaller number represents a higher degree of accuracy. In the study, researchers found that adding more oscillators significantly improved the model's results.

In the CIFAR-10 test, scores ranged from an FID of 11.01 with 1,024 oscillators to 8.76 with 4,096 oscillators. In the more demanding ImageNet 64x64 test, a pool of 6,656 oscillators achieved a score of 8.41 FID, while 16,384 oscillators clocked in at 6.74.

These results are comparable to those of early image generation models, including Google's pioneering BigGAN and OpenAI's iDDPM, which paved the way for its more modern DALL-E tool. However, the study authors stressed that the results "should be read as reference points rather than strictly identical measurements."

"We view Un-0 as a promising first approach with quality that overlaps with that of several established image generation families when they were first introduced to the community," company representatives said in the technical blog post. "Un-0's quality matches where today’s leading generative methods began. Conventional generators are still stronger on absolute quality and parameter efficiency — closing that gap with new algorithms and model architectures is the work ahead."

The scientists released the model weights — the internal mathematical parameters that the model alters as it learns — as well as training and ablation scripts — specialised code files used to build and test the system — allowing other researchers to test the models and run their own simulations. They hope to close the gap with new algorithms and models.

"Taken together, Un-0's system of coupled Kuramoto oscillators offers the promise of learning with physical dynamics at a scale that's beyond what has been done before," they said in the technical blog post. "Un-0 points in the direction of the opportunity for a new computer that exploits physics to achieve our top-line goal of energy efficiency."

Much of the discourse around artificial intelligence (AI) focuses on grand ideas such as the rise of a hypothetical artificial general intelligence (AGI) and superintelligence. Speculation swirls around the likelihood that the technology will thin out the job market, or even precipitate the death or evolution of human creativity. We haven't focused as much on the multitude of subtle yet hugely consequential ways in which AI is reshaping the social fabric of our society, and how we collectively imagine the future.

That's the argument sociologist and AI researcher Mona Sloane, an assistant professor of data science and media studies at the University of Virginia, puts at the center of her new book, "Predicted: How AI Is Restructuring Social Life" (University of California Press, 2026). Whether we consider email filtering, prediction markets or social media platforms, AI systems are embedded in the heart of how we interact with the digital world. Indeed, AI is so ubiquitously integrated into everyday interfaces that it's given rise to a new kind of "prediction logic" that makes assumptions about who we are and how we are likely to behave.

In this excerpt, Sloane compares the AI technology we use today with the oracles of ancient Greece, framing it as an omnipotent presence that has moved to organize society through the prism of prediction models. This in turn affects how we learn, live, love and even picture the future.

We live in a world of oracles. These oracles constantly feed us predictions that shape our social lives — how we socialize, love, work, gain access to resources. Like in ancient Greece, predictions occupy a prominent role in our society. We consider our oracles so mighty that their predictive power rules over the fate of whole economies and even geopolitical constellations. Where the oracle is, there is the center of the world.

But unlike in ancient Greece, our oracles aren't high priestesses delivering divine prophecies. They are artificial intelligence (AI) systems melted into the infrastructure of everyday life. Today, it is nearly impossible to evade the grasp of AI predictions. I voluntarily and involuntarily use AI on a constant basis: by using email providers that build on the predictive properties of AI for spam filters, by conducting online banking and getting enrolled into AI-automated fraud detection, or by using generative AI for supporting administrative chores. It has become part of how I experience the world.

It can be a relief when it helps me do things I dread or am bad at, such as produce a spreadsheet template I desperately need, help streamline language produced by different authors for a report, or generate a specific image for a presentation. Often, I must intently handhold the AI, checking and fixing its outputs. And sometimes, with deep frustration, I give up and start all over to complete my task manually.

The omnipresence of AI prediction can make it easy to think of these systems as inevitable, quasi-natural phenomena we are subject to, rather than a part of. But they are quantitative concepts that arise from social agreements about how we ought to capture and interpret the world around us.

"Quantitative concepts are not given by nature: they arise from our practice of applying numbers to natural phenomena," wrote Rudolf Carnap, a logician and professor of philosophy of science, in 1966. His point was that numbers can be useful, because they allow for information to travel more easily across contexts, as a sort of language. They also make mathematical predictions possible.

To him, this was first and foremost useful for engineering modern life: A quantitative language allows for the articulation of quantitative laws that, in turn, facilitate the routine generation of mathematicized predictions, particularly in the realm of physics. Being able to predict how energy, compounds, and materials will behave in certain configurations is the reason humans were able to build the conveniences of airplanes, cars, and telephones. For Carnap, predictions were simply instrumental in this way.

Today, almost 60 years later, this pragmatic approach to mathematical prediction has been turned on its head by AI. Prediction is no longer just a handy tool in physics or engineering. The promises of AI's oracular power have turned prediction into a logic for structuring social life. This is a dangerous proposition. It implies that AI is always necessary or even inevitable and diverts attention from the social forces shaping ideas around this technology in the first place.

AI systems are not natural phenomena that happen to us. They are collective expressions of society. As such, they are not just a hype or a deception concocted and executed by global tech elites. They indicate a wider shift in how we imagine and enact society. Many critical discussions of AI characterize this phenomenon chiefly as heightened surveillance and capitalist extraction. But this is a myopic diagnosis. AI's most powerful effect is the subtle yet comprehensive recalibration toward prediction as a guiding principle for organizing society. In this book, I call this phenomenon the prediction paradigm.

AI is something that we do as part of going about our lives and participating in society — it is social infrastructure, affecting how we relate to one another and how we act in public and in private. Like all infrastructures, AI allows resources and ideas to flow in certain directions, but not others. AI uses data from our collective past to predict our individual future. And because AI deals in futures, it solidifies a linear time regime that hardens our social commitment to causality: The past always predicts the future. The problem of AI is not the rise of intelligent machines, but the extraordinary social significance ascribed to this linearity, fetishizing the future and leaving little room for deliberations about what (other) futures may be possible or we may want.

Reprinted from Predicted: How AI Is Restructuring Social LIfe by Mona Sloane, courtesy of the University of California Press. Copyright 2026.

The massive hardware demands of artificial intelligence (AI) applications are stretching the physical and structural limitations of semiconductors. But researchers have engineered a three-dimensional silicon chip that they propose as the solution.

In a new study published May 27 in the journal Nature, scientists found a way to cram more computing power into a chip by stacking silicon circuits in multiple layers in a way that doesn't impact performance.

Stacking chips vertically, known as 3D integration, is more efficient than traditional 2D chips, where silicon circuits are spread across a single surface. This is because stacking shortens the distance that data has to travel and reduces the power required for data transmission.

The researchers' 3D chip uses ultrathin silicon membranes and low-temperature manufacturing techniques to overcome the challenges of current chip architectures.

"Our method is not only easier to implement with lower cost, but it has several advantages over previous approaches to stack silicon wafers," Qing Cao, first author of the study and a materials science and engineering professor at the University of Illinois Urbana-Champaign, said in a statement.

Extending Moore's law

Since the 1960s, ensuring that electronics can handle more demanding applications has meant making transistors smaller so more can be packed onto a single chip. But, as Cao pointed out, doubling the number of transistors every couple of years — a principle known as Moore's law — is becoming less feasible.

"If you look at the actual size of transistors, they're not getting smaller, especially in terms of their contacted gate pitch," Cao said in the statement — defined as the combined width of one transistor gate and the space needed to separate it from the next.

"This is because we're becoming limited by the intrinsic material properties of silicon and the fundamental rules of quantum mechanics. If we're going to keep up the trend of increasing processing power of our microprocessors, we have to start thinking beyond just squeezing more devices on a single surface."

The researchers think vertical integration across multiple layers is the best way to guarantee that engineers can continue to adhere to Moore's law, because this approach creates room for more transistors on a chip.

"Today it takes six microelectronic devices called transistors on a single plane to store one bit of information," Cao explained, suggesting that just like in a densely populated city, the only way to solve overcrowding is to build upward. "You get the same functionality, but the spatial footprint is reduced while making communication between layers faster and more efficient."

Getting around the heat problem

Stacking is nothing new, of course, but vertical integration — building layers directly on top of one another — can create thermally dense packages. In the study, the researchers noted that the fabrication of high-quality silicon chips demands temperatures up to 1,832 degrees Fahrenheit (1,000 degrees Celsius).

However, once the first chip layer has been completed, the metal wiring introduced to connect further layers can be destroyed by such high temperatures. As a result, the "thermal budget" — the maximum amount of heat that can be endured before degradation starts to occur — for any additional layers is 752 F (400 C), said Cao. This can result in performance and reliability issues.

When creating 3D stacked silicon chips, manufacturers have sought to avoid this problem by using alternatives to single-crystalline silicon for the upper layers, according to the researchers. These materials include amorphous and nanocrystalline metal oxides, carbon nanotubes and polycrystalline silicon, but they can lead to performance and reliability issues, the scientists said in the study.

To overcome this challenge, Cao and his team adopted an approach called "monolithic integration" — a process in which all chip components are fabricated on a single piece of substrate, as opposed to making them separately and then bonding them together later.

To build each chip, the researchers created ultrathin silicon nanomembranes that they then transferred, using a roll laminator, onto a substrate containing the bottom layer.

The maximum temperature required to generate a strong bond using this method was just 392 F (200 C) — five times less than the heat normally required. The membranes they transferred were also just 10 nanometers thick or less — about the size of a protein — compared with the approximately 500-to-700-micrometer (500,000 to 700,000 nanometers) thickness of a typical wafer. Because they are thin, these membranes are mechanically flexible to conform to the underlying surface, Cao added.

The result of this process was a 3D chip with three layers, each containing 625 transistors. This pales in comparison to the billions of transistors that can be crammed onto chips already on the market, but the researchers believe their technology boasts power efficiency benefits. The electrical current that can flow through the chip has proved to be at least three to four times greater than that of monolithic chips made from alternative materials.

The big question is whether their 3D silicon chip can make the leap from the laboratory to commercial applications. While the research demonstrates the potential of a chip comprising three stacked layers, the scientists suggested that plenty more layers can be added in future iterations.

Can you match these ancient devices to their pictures? Find out with our computing quiz!

A four-legged robot has learned to change the way it runs while navigating forests, staircases and obstacle courses. — seamlessly switching between a steady trot and a faster bounding gait without instructions from a human operator.

The 100-pound (45 kilograms) robot, called KAIST HOUND, uses cameras and lidar to scan the ground ahead, then selects an appropriate gait and adjusts its movements in real time. In outdoor tests, it crossed a 0.7-mile (1.1- kilometers) university campus route and a 0.2-mile (0.3 km) forest trail strewn with roots, logs and slippery leaves.

The researchers described the robotic framework on July 15 in the journal Science Robotics.

Changing gait

Animals naturally change their gait depending on their speed and surroundings. A dog might trot carefully across uneven ground, for example, before bounding over a fallen branch. Reproducing this adaptability in robots is tricky because different movements are often controlled by separate, highly specialized coding systems, and transitions between them can cause a lag that drives the robot to stumble.

To overcome this issue, researchers developed a special training framework called action pretrained transformer–based reinforcement learning (APT-RL). This is an artificial intelligence (AI) training system that first studies many examples of actions, uses a transformer to understand patterns across those actions, and then improves through rewards and penalties.

The training began with a simple, two-dimensional computer model of the robot. Using trajectory optimization — a technique that calculates physically workable movements for the robot — the team generated 180,000 short trotting and bounding sequences, including the joint forces the robot's legs need to perform. The dataset represented about 15.5 hours of movement but took only around eight minutes to produce.

During reinforcement learning — a machine learning technique where AI learns to make the best decisions by engaging with a particular environment through trial and error — an AI system then learned how to select and modify those skills while negotiating simulated stairs, stepping stones, hurdles, gaps and rough ground.

In digital simulations, the robot dog was not limited to copying its prerecorded movements. It could also make corrections for three-dimensional terrain and unexpected situations, such as jumping over a log — a behavior that wasn't included in the original, flat-ground training data.

Finally, the researchers configured the system to include the robot's depth camera and lidar scanner in the simulation.

In one indoor test, HOUND bounded across an obstacle 2 feet (60 centimeters) high while briefly achieving 9.5 mph (15 km/h). It also jumped down a three-step staircase. The robot generally chose trotting at lower speeds on irregular ground, while bounding became more common at higher speeds or when it encountered larger steps, hurdles or gaps. The AI system that could select either gait performed more consistently across the different simulated environments than the version restricted to trotting or bounding alone.

The researchers suggest the technology could eventually help robots navigate disaster zones or other places inaccessible for wheeled machines. However, the current framework only allows two gait choices and mainly handles forward movement. Rapid turning, sideways motion and other behaviors like crawling remain future goals for the research team.

Consider the following scenario. Suzy is 63, recently retired, and trying to decide when to start receiving Social Security and how to manage her retirement savings to minimize the tax hit.

She opens an AI chatbot, types in the details and gets a calm, well-organized and confident answer: Claim now, convert this much, here is the reasoning.

The chatbot sounds authoritative and even shows its work. So Suzy follows its guidance and never calls a financial planner. Maybe the advice was fine. But maybe it quietly ignored the fact that Suzy's spouse is younger and in poor health, which can flip the Social Security math. It also may have overlooked that the retirement savings plan conversion it suggested would push Suzy into paying higher Medicare premiums two years later.

Suzy won't find out for a long time, if ever, whether this guidance was right for her. And the AI will never call back to say it was unsure.

Suzy isn't an exception. AI chatbots have entered everyday life with remarkable speed: A 2025 Pew Research Center survey found that 34% of U.S. adults and 58% of those under 30 have used ChatGPT, roughly double the share two years earlier.

A growing number are asking AI about money, and some are getting burned. According to a 2025 survey of 2,000 U.S. adults by Pearl.com, a professional services platform, 19% said they lost more than $100 by following financial advice from an AI chatbot. Among Gen Z investors, that figure rose to 27%.

These aren't hypothetical risks. People are already paying for answers about their money that are confident — and wrong.

As a finance professor who has been closely watching the spread of AI into personal finance, this is the part of the AI story that worries me most. And it's not the part you usually hear about.

We argue about AI the wrong way

There are two seemingly opposite complaints about AI. One is that people trust it too much, treating a chatbot like an oracle, a tendency researchers call algorithm appreciation. The other is that people don't trust it enough and dismiss its useful tools, a tendency known as algorithm aversion.

I argue these are actually two sides of the same coin, and what decides which side you see is whether you can tell when the AI is wrong.

When an AI fails in an obvious way, you notice and lose confidence. So you're more likely to seek a professional or another human you trust sooner than you otherwise would. That is the safe failure.

The dangerous failure is the opposite. The answer is fluent, confident — and wrong. You have no way to catch it, so you keep managing the problem yourself long past when you should have asked for help.

The trouble is that with money, the second kind of failure is the common kind.

When you mistake fluency for accuracy

Three things make financial advice especially treacherous for AI.

First, fluency is not accuracy. People naturally read a confident and well-articulated answer as competent. But how polished an answer sounds tells you almost nothing about whether it fits your situation or the accuracy of the proposed solution. A chatbot can be word-perfect and still be wrong about your taxes, because your taxes depend on details it never asked about.

Second, AI is least reliable exactly where the stakes are highest. AI tools are good at routine and general topics: what a Roth IRA is, how compound interest works, the difference between a stock and a bond.

But financial life is full of rare, complicated, one-time decisions: exercising stock options, understanding the alternative minimum tax, making required, minimum 401(k) distributions, deciding on a Social Security strategy as a couple, drawing up a divorce settlement.

I made a similar argument three years ago about AI trading on Wall Street. Because market crashes are rare, there's little data for AI to learn from, so it can be most confident exactly where it is least informed.

That worry hasn't faded. Market watchers now caution that AI trading bots are creating fresh financial risks, and that same blind spot applies to your personal finances. Researchers call this uneven competence a "jagged frontier" — reliable with common cases but unreliable for unusual ones. And in finance, the unusual cases tend to be the expensive ones.

Third, you often can't check the work. Financial advice is what economists call a "credence good," like a mechanic's diagnosis or a doctor's recommendation. You often can't tell whether the advice was good, sometimes for years. A mistaken tax move may not surface until an audit. A bad 401(k) drawdown plan may not bite until the stock market slumps. Without quick feedback, the wrong-but-confident answer never gets corrected.

This is why the Pearl numbers above are probably an undercount, since they capture only losses people noticed.

The quiet failure is the one to watch

Notice that the real harm in Suzy's story isn't a single dramatic mistake. It's that a confident answer made Suzy feel no need to call a professional, so the call never happened.

The danger is not so much that you act on bad advice but that you never seek good advice. The smoother and more reassuring the tool, the easier it is to stay in do-it-yourself mode past the point when you need outside help.

Who's most at risk? In a study of a large robo-advising platform in India, co-author Vishaal Baulkaran and I found that its users skew young, are predominantly male and tend to be smaller retail investors and professionals. And new account sign-ups rise during periods of high market volatility.

In other words, the people leaning hardest on automated advice match that 27% figure among those Gen Zers who lost more than $100 while using a chatbot for financial advice. They reach for it just when markets turn turbulent and a wrong move is most costly.

There's also an incentive worth naming. In my new analysis, I argue that a tool that earns its revenue by holding your attention has a reason to sound confident and helpful: Confidence keeps you on the platform. The catch is that the user it retains that way is sometimes the one who should have been handed off to a human.

A system tuned to keep you engaged isn't the same as one tuned to protect your financial future, and the two can point in different directions. The disruption is already underway, as wealth managers face what Bloomberg has called a chatbot reckoning. A single, new AI tax tool recently sent wealth management stocks sliding as investors bet that automated advice will eat into the business.

How to be smart about using AI

These findings don't mean that people should avoid AI for money advice. Used well, these tools are a valuable and free financial educator.

This is also not to say that a financial adviser always has the right answers. As with finding any kind of specialist, it's important to do research first and make sure they meet the kind of criteria laid out by the Consumer Financial Protection Bureau. Fee transparency is also crucial.

But if you do turn to AI, the skill is knowing where to draw the line.

Treat AI as a starting point, not a verdict. It's excellent for learning concepts, drafting questions and getting oriented before a meeting. It can teach people the vocabulary to have a smarter conversation with an expert.

But watch out for the signals that you have left its comfort zone and entered the territory where AI is weakest and a confident answer is least trustworthy. The red flags are large dollar amounts, tax consequences, anything irreversible and anything that turns on the specifics of your situation rather than a general rule.

Estate questions, the drawdown of retirement savings, strategies for claiming Social Security benefits, business structure and major one-time transactions all belong in this category. Those are the decisions that call for bringing in a human, such as a certified financial planner.

And remember, confidence isn't competence. When the answer about your money sounds most polished and most certain, that's not a reason to relax. On the hardest questions, that smooth confidence is exactly the signal that you should pick up the phone and talk to an expert.

This edited article is republished from The Conversation under a Creative Commons license. Read the original article.

Fast charging seems to be almost everywhere. Many smartphones can go from nearly empty to more than 50% charged in about half an hour, while some electric vehicles can add hundreds of miles of range during a quick charging stop.

But batteries aren't perfect; their capacity degrades over time. Given that fast charging delivers more power in a shorter amount of time, does fast charging damage batteries?

Scientists say the answer is yes, but it's more complicated than you might think. Fast charging can accelerate some types of battery degradation, but modern batteries are designed with safeguards to help limit the damage.

How does fast charging work?

Rechargeable lithium-ion batteries — the most common battery type in the world — work by moving lithium ions between two electrodes called a cathode and an anode. During charging, lithium ions travel through the battery and are stored in the anode until the battery is used again.

The main difference between fast charging and regular charging is how quickly those ions move. Compared with regular charging, which can take hours, fast charging can refill a battery in an hour or less.

"Regular charging applies a lower current, allowing lithium ions to intercalate [move into microscopic holes] into the anode gradually, which generates little heat and causes minimal mechanical stress," Zhiyuan Jiang, an associate professor in the Department of Chemical Engineering and Technology at Xi'an Jiaotong University in China, told Live Science via email. "Fast charging increases the current [and] power significantly to shorten charging time."

Not all batteries are designed for fast charging. A battery's ability to handle high charging speeds depends on its materials, internal structure and battery management system, Jiang explained. Fast-charging batteries often use specialized electrode materials or thinner electrodes and electrolytes that allow the lithium ions to move more easily. Manufacturers may also redesign the battery's internal architecture to reduce resistance and heat buildup.

Stanislaw Zankowski, a battery researcher at the University of Oxford, compared the process to traffic moving through a city.

"You could think about charging a battery as transporting people through roads, intersections and buildings," Zankowski told Live Science. "Fast charging is really a question of how efficiently you can move all that traffic without creating bottlenecks."

What type of damage could fast charging cause?

All lithium-ion batteries lose capacity over time, even when they are treated carefully. But fast charging can speed up some of the chemical processes responsible for that aging.

One of the biggest concerns is a process called lithium plating. During rapid charging, lithium ions may not have enough time to settle properly inside the anode. Instead, some lithium can accumulate as metallic deposits on the electrode's surface. These deposits can reduce the amount of lithium available to store energy, thereby lowering the battery's capacity. In extreme cases, the lithium can form needle-like structures called dendrites that puncture internal battery components and create safety hazards.

Fast charging can also generate more heat. Heat is a natural byproduct of electrical resistance in the battery. The faster a battery charges, the more heat it produces.

"For charging a small battery with a small current, that amount of heat will be also relatively small," Zankowski said. "So, it's not really a safety problem, but as we increase the size of the battery, the amount of current that we'll be pushing during charging and the amount of heat will also increase quite a lot. And as a result, we can't really charge larger batteries as quickly, mostly because of the safety margin."

Higher temperatures can accelerate chemical reactions that gradually degrade battery materials. In some extreme cases, overheating can cause batteries to swell, catch fire or even explode ‪—‬ a process known as thermal runaway.

Fortunately, most modern smartphones, laptops and electric vehicles contain sophisticated battery management systems that monitor voltage, current and temperature during charging. That's why you may get a heat warning from your smartphone if you leave it in the sun. If temperatures climb too high, these systems automatically slow charging to protect the battery.

Best tips for battery life

So what's the best way to protect a battery while charging it quickly?

Both Zankowski and Jiang emphasized that temperature is key. It's best to avoid charging devices in hot environments, such as inside a parked car or in direct sunlight. Extremely cold temperatures can also be harmful because they make it harder for the lithium ions to move through the battery.

"Ideally, the temperature range should be around 20 to 25 degrees Celsius [68 to 77 degrees Fahrenheit] for charging," Zankowski said. "So, just like a comfortable temperature for a human being, right?"

Experts also recommend that you avoid keeping devices like laptops constantly plugged in, as this can degrade battery performance. Jiang suggested implementing the "shallow charge, shallow discharge" habit.

"Keep your battery between 20% and 80% for daily use," he said. "It is not necessary to charge to 100% every time."

Can you match these ancient devices to their pictures? Find out with our computing quiz!

Researchers in China have announced a radical sodium metal battery (SMB) design that can fully charge in just four minutes and will retain its capacity for years of use.

SMBs are a form of ultrafast-charging, stable batteries that scientists say could one day be a cheap alternative to today's lithium-ion (Li-ion) batteries, which rely on geographically concentrated metals and easily catch fire. SMBs also differ from sodium-ion (Na-ion) batteries in that they use a metallic sodium anode rather than a graphite or hard carbon anode.

However, SMBs remain largely theoretical because they are prone to a type of degradation known as dendrite formation. This is when the sodium ions passing through the electrode deposit onto the highly reactive, pure-metal sodium anode in spiky, stalagmite-like structures. Over time, this forms a bridge between the cathode and the anode, short-circuiting the battery.

Dendrite formation is especially common in sodium batteries because sodium is a highly reactive metal. When charge runs through a Li-ion, Na-ion, or sodium metal battery, the anode always reacts with the electrolyte to form an oxide layer known as the SEI. This is typically 10 to 50 nanometers thick — about as wide as a small virus — but generally harmless. But with sodium, the SEI often cracks, forming bumps that attract sodium ions, which pile into dendrites.

Now, researchers say they have solved this issue by using a tough, quasi-solid gel electrolyte — dubbed Sn-FB QSE — which strengthens the battery against punctures and provides a semisolid internal structure that prevents dendrites from forming. They outlined their findings in a study published May 21 in the journal Nano-Micro Letters.

To confirm the longevity of this approach, the scientists charged and discharged the battery for over 6,000 hours without dendrites short-circuiting the battery. They also noted that when they charged the battery from zero to 100% capacity in just four minutes, it retained electrical charge, measured in milliampere-hours per gram (mAh g^–1), of 80.1. This is the equivalent of around half that retained in Li-ion batteries.

When charged at a slightly slower rate of zero to 100% in 20 minutes, the battery retained 90% of its charge capacity over 2,000 cycles — matching the theoretical limits for Li-ion batteries, the scientists said in the study. This slower speed lowered the cost and improved the safety.

This is notable because the scientists achieved this in the new battery while still charging it quicker than Li-ion batteries can be charged. This is relevant because charging speed remains a sticking point for battery deployment in electric vehicles (EVs). The fastest charging EV today is the BYD Denza, which the Chinese automaker says can go from 10-70% in just five minutes. But this requires highly specialised, 1MW proprietary chargers.

Most EVs charge much slower — Tesla representatives say its Model 3 can recharge from 10-70% in approximately 15 minutes using Tesla’s own 250kW flash chargers, but representatives from the EV routing platform Zapmap say the same vehicle will take 90 minutes to charge to 80% on 50kW chargers.

Indeed, most batteries used for modern technologies, such as smartphones and EVs, are Li-ion. However, Li-ion batteries are expensive to produce because they contain the hard-to-obtain metals lithium and cobalt, and they are prone to catching fire.

Increasingly, battery manufacturers are looking to bring Na-ion batteries to commercial scale because they are cheaper and safer. However, they are heavier and larger than Li-on batteries.

SMBs are the focus of intense research because they theoretically combine the best of both types of batteries. Because SMBs use a sodium anode, rather Na-ion batteries that use graphite or hard carbon anode, they are lighter and cheaper to produce and therefore much more comparable to Li-ion in terms of size and weight. They are also safer because they operate using sodium ions, which are bulky and cannot flow to breaches in a battery wall fast enough to cause thermal runaway. This is the self-sustaining chain reaction that causes batteries to ignite when damaged.

If the issues of dendrite formation and stability at lower temperatures can be resolved, replicated and scaled, SMBs could reshape the economics of battery deployment over the next decade, the scientists said.

SMBs could be excellent choices for EVs in public transport or within commuter cars, the scientists belive, because although they have lower ranges than Na-ion and Li-ion vehicles do, they charge faster. However, they won't be available for some time, either in vehicles or smaller devices like consumer electronics.

That's because devices like smartphones are subject to harsh temperature changes that affect the internal chemistry of batteries that rely on gel electrolytes. The research must first be replicated before manufacturers feel comfortable using pure sodium metal in place of well-understood graphite configurations.

Engineers have designed a waterproof "diving suit" for cyborg cockroaches that enables the hybrid insects to survive and roam underwater for up to three hours. This function expands the capabilities of cyborg insects and could one day be deployed in disaster zones, according to the team.

A built-in oxygen generator and silicone tubes deliver the gas directly to a cockroach's breathing holes, known as spiracles. The design is adapted for use in low-oxygen conditions as well as submerged environments, the researchers said in a new study published June 29 in the journal Nature Communications.

"Our approach combines a soft waterproof shell with a simple yet reliable chemical oxygen generator," study co-author Shinjiro Umezu, a professor in the School of Creative Science and Engineering at Waseda University in Japan, said in a statement. "This allows the insect to retain its natural mobility while being protected from an environment that it cannot normally survive in."

Cyborg insects are living insects that have been fitted with electronic controllers that guide their movements. Researchers have previously used them in search-and-rescue operations to access and investigate hard-to-reach areas; for example, they were used in rescue efforts after the devastating magnitude 7.7 earthquake in Myanmar in March 2025 that killed at least 3,700 people and injured 4,800 more. The advantage of cyborg insects over tiny robots is that the former employ insects' muscles to move, whereas the latter rely on high-power batteries that consume energy and can run out of steam.

The cyborg insects deployed in Myanmar were developed in the laboratory of Hirotaka Sato, senior author of the new study and a professor in the School of Mechanical and Aerospace Engineering at Singapore's Nanyang Technological University.

Sato has spent more than a decade pioneering cyborg insect technology. He and colleagues hope the new diving suit will extend cyborg insects' operational range to include flooded and partially submerged areas in disaster zones.

The suit consists of a flexible shell, four silicone tubes that attach to the spiracles and a transparent, 3D-printed oxygen tank. To make the tank produce oxygen, the researchers sprinkled manganese dioxide onto a highly absorbent sponge inside the tank. They then injected a small amount of diluted hydrogen peroxide, which breaks down slowly in the presence of manganese dioxide to produce oxygen. Finally, the team sealed the tank with ultraviolet adhesive to prevent leaks.

"The key engineering challenge was to build a system that was small, light and flexible enough for the insect to wear, while still producing enough oxygen for long-duration underwater movement," Umezu said.

The silicone tubes send oxygen straight into the thoracic spiracles, while the abdominal spiracles, which are lower down the insects' bodies, take in the oxygen contained in the suit.

"Our new insect diving suit works like the oxygen tank used by human divers," Sato said in the statement. The silicone tubes can be attached and removed without pain or harm to the insect, the researchers added.

The researchers tested the suit on a cyborg Madagascar hissing cockroach (Gromphadorhina portentosa), which they placed in a water tank and later sent into a plastic tube that simulated submerged and low-oxygen environments.

The suit enabled the cockroaches to roam underwater for up to three hours, raising the prospect that cyborg insects, including locusts and beetles, could one day be used to inspect flooded pipes, drains, tunnels and other hard-to-access places.

Next steps include improving the diving suit to potentially include sensors and a navigation system; and testing the design in simulated disaster environments, according to the statement.

Using a quantum computer alongside a supercomputer, scientists have developed a breakthrough pathway for modeling the physics inside a fusion reactor. The world-first experiment could help clear a path to developing clean, abundant nuclear power and solving the global energy crisis, the researchers said.

Using hybrid quantum computing and artificial intelligence (AI) methods, scientists with IBM and Oak Ridge National Laboratory (ORNL) have blueprinted how to make tritium, an extremely rare isotope of hydrogen that's critical to the fusion process.

Although their research — uploaded June 29 to the preprint server arXiv — has not been peer-reviewed, the researchers say it's the first time that different kinds of computing elements have come together to propose the most effective way to create this material.

Fusion reactors are experimental power sources that create energy by fusing atomic nuclei. The heat produced in the subsequent nuclear reaction is then harnessed as energy. This method produces no carbon byproducts or long-lived radioactive waste, making it one of the cleanest potential forms of mass energy production.

It's projected that, at scale, a single fusion reactor could produce about 4 million times as much energy as a coal-burning facility and around four times the amount of energy as a modern nuclear fission reactor.

Current attempts at building a viable fusion reactor have resulted in numerous laboratory experiments that prove the technology works, with magnetic confinement reactors, such as tokamaks, widely considered the front-runner. But many engineering challenges remain before the first commercial reactors could come online.

Turning seawater into fuel

The base fuel for nuclear fusion reactors is a hydrogen isotope called deuterium, which is commonly found in seawater. It's estimated that there are 33 grams of deuterium in every cubic meter of seawater.

But deuterium is only half of the equation. Nuclear fusion also requires tritium — a heavier hydrogen isotope — and the fusion released from just 1 gram (0.04 ounces) of deuterium-tritium fuel equals the energy from about 2,400 gallons (9,100 liters) of oil, according to the U.S. Department of Energy.

Unfortunately, tritium, a radioactive isotope, is extremely rare; only 44 pounds (20 kilograms) of it is produced on Earth each year, and its 12-year half-life makes it difficult to use in nuclear power plants.

Instead, scientists must painstakingly produce tritium in nuclear reactors by bombarding lithium atoms with neutrons. It's then superheated and bound with powerful magnets into a whirling ring of plasma within a tokamak, a special fusion chamber designed to shape and heat plasma using magnetic fields.

Scientists add more deuterium and then bash the tritium and deuterium together, causing them to fuse into helium. The force of this reaction creates heat that's converted into energy.

The current bottleneck lies in creating enough tritium to sustain fusion long enough to produce energy. But modeling the particle physics and chemical reactions involved in the tritium-creation process has proved beyond the capabilities of classical supercomputers.

In the new study, however, scientists say they have addressed this bottleneck by simulating nine molecular configurations of a liquid salt that contains fluorine, lithium and beryllium (FLiBe) — one of the leading candidate materials for extracting tritium.

This is the first time quantum computers have been used to model reactions inside a fusion reactor. If perfected, FLiBe could provide a near-limitless source of fuel for nuclear fusion reactors, they said, but the chemistry involved is incredibly complex.

Demystifying complex chemistry

A "blanket of molten salt" made of FLiBe is wrapped around the nuclear reaction inside a fusion reactor, IBM researchers told Live Science. This provides both a fuel source and a thermal shield for the device.

To create enough tritium, the researchers had to calculate the physics involved while a process called "neutron bombardment" constantly altered the blanket's chemistry. Designing a salt that holds up under competing demands and keeps releasing tritium is a key problem in building this kind of reactor.

"If tritium grabs onto fluorine in the salt, it forms tritium fluoride, which is corrosive and stubborn to remove," the researchers explained. "If it binds to another tritium atom to form a gas, it bubbles out on its own. Predicting which way the reaction goes means modeling the interaction between tritium and the salt with high precision and accuracy that is challenging for classical methods."

Because no ordinary computer can perform the necessary calculations, the team used a combination of AI running on the Frontier supercomputer at ORNL, alongside quantum computing algorithms running on an IBM Quantum Heron quantum processing unit (QPU) in New York. The resulting workflow demonstrated a proof of concept for offloading complex chemistry computations to a quantum computer.

That workflow relied on a technique called wave-function-based embedding, which fragments the calculation into easier-to-calculate clusters, the scientists said in the study. They used classical computers to solve the smaller clusters and passed off the more difficult chunks to a quantum computer. The classical computers then stitched the molecule back together.

This is a method that study co-author Kenneth Merz, a biochemist and principal investigator at Cleveland Clinic Research, pioneered in previous research. Earlier this year, in collaboration with IBM and the Japanese national research institute RIKEN, he used quantum computers to calculate the structure of a 12,635-atom protein.

Fusing quantum and AI

In the new study, the researchers tested their model against known molecular configurations that were already solved by a nonhybrid classical system and determined that the accuracy was maintained with the addition of quantum computations.

This proof of concept should serve as a direct pathway for scaling the models used to predict tritium production within fusion reactors, potentially solving what may be the biggest hurdle to large-scale fusion energy production.

The broader workflow the scientists outlined in a technical blog post involved three stages. First, AI agents proposed and screened many candidate salts from the ORNL database, and for each candidate, calculations estimated various qualities in the tritium breeding process, including how much fuel the salt would make under neutron bombardment.

The most promising salts then went to a supercomputer, which modeled them atom by atom, using the density functional theory (DFT) process to approximate how a molecule's electrons would arrange themselves. These are expensive simulations, so the scientists used "AI stand-ins" trained to reproduce the physics to run them fast enough to be useful. The third stage brought in the quantum computer to figure out where the tritium would bind, which is a shortcoming for DFT.

In the future, the research team will model larger molten-salt systems and study more molecular configurations before evaluating whether AI can slash the time it will take to find a promising molten-salt material.

The wider aim, the scientists told Live Science, is to build a reliable computational pathway for fusion-materials discovery that can help researchers predict how well a blanket material breeds tritium, whether that tritium can be recovered, and how the material may perform in the extreme environment of a fusion reactor.

Can you match these ancient devices to their pictures? Find out with our computing quiz!

When you click to enter a website or try to log in or fill out a form, you may be asked to identify motorcycles from a grid of grainy images, decipher a string of convoluted characters, or click a box that states "I am not a robot."

These tests are called CAPTCHAs, which stands for "Completely Automated Public Turing test to tell Computers and Humans Apart." As their name suggests, they are meant to help a website distinguish if an action is coming from a human or a bot, since the aforementioned tasks are theoretically easy for a human and difficult for automated software to perform. This, in turn, blocks bots from spamming comments, downloading files, taking over accounts, or executing any other action on a website.

But as computer models increasingly gain the ability to solve CAPTCHAs, thanks to advancing artificial intelligence (AI) — and with puzzles getting weirder and more difficult for humans to complete — does this mean CAPTCHAs are still useful?

CAPTCHAs were introduced in the late 1990s to address "a very simple, but very difficult problem," Andreas Plesner, a computer scientist at ETH Zurich, told Live Science. "If I don't interact with a person physically … is it a computer? Or is it human?" Some of the first CAPTCHAs, still common on websites today, were composed of distorted text, since text-reading software at that time had trouble interpreting warped words.

But over time, text-reading software improved and new types of CAPTCHAs were developed. For instance, reCAPTCHA, one of the most popular CAPTCHA services, has an image-based test that asks users to identify objects such as traffic lights, motorcycles or bicycles from a grid of Google Street View photos. This was developed after Google acquired the service in 2009.

"The bet was that recognizing objects in messy, real-world photos was still a uniquely human skill," Ng Chong, chief of information technology and director of United Nations University's Campus Computing Centre in Tokyo, told Live Science in an email.

As time went on, CAPTCHA design continued to advance. In 2014, Google came out with reCAPTCHA v2, which analyzed computer mouse behavior by asking people to click a checkbox to test if a user was human. If the behavior is deemed suspicious, determined by factors like how a user interacts with the site beforehand or the timing of their click, the street-image grid pops up as an additional puzzle.

However, more recently, technology has improved to a point where image recognition is no longer a human-specific skill. As early as 2016, researchers found that low-cost deep learning technologies could solve reCAPTCHAv2 around 70% of the time. By 2024, Plesner and his colleagues developed an AI model that could solve the puzzles correctly 100% of the time. Earlier in 2026, Chong noted that he built a tool that could mimic human-like browsing behavior and sometimes bypass reCAPTCHA v2 without triggering the image grid at all. When the grid was triggered, the tool used AI to solve it within a few tries.

"When both the challenge and the behavioral layer are defeated by commodity tools running on a single laptop, the fundamental premise of CAPTCHA, that there are tasks humans can do but machines can't, stops holding," Chong wrote.

Looking ahead

So does that mean CAPTCHAs are completely obsolete? Not quite. Although the model Plesner and his colleagues developed breezed past reCAPTCHAv2, "there were a lot of the safety measures that were not tied to being able to solve it, but more tied to how you solve it," he said. For example, while conducting their research, Plesner noted that his team used a virtual private network (VPN) that changed IP addresses for each test, since a single IP address with a high volume of solved CAPTCHAs faced tasks with increasing difficulty, or got blocked entirely.

Modern CAPTCHAs focus on these background clues and tactics, rather than the puzzle itself. This includes Google's reCAPTCHA v3, Friendly CAPTCHA, hCAPTCHA and Cloudflare's Turnstile, among others, which run without sending a puzzle at all. They instead look at whether the action is coming from a real attested device (rather than from automated code), whether an IP address has had a high volume of automated requests in the past, how a user navigates a webpage, what the user's cookie history is, and a slew of other factors to determine possible malicious intent.

As the tug-of-war continues, CAPTCHA puzzles are still widespread. After all, they’ve been the status quo for decades, are easy to set up and are relatively cost-effective, Chong said. But these tasks have some other drawbacks. Although bots can increasingly solve the puzzles with ease, CAPTCHAs can be a headache to get through for humans and can be seen as discriminatory against those with disabilities, notably visual disabilities, as a researcher noted in a 2022 conference paper.

The crescendoing complexity of CAPTCHA puzzles has even been the subject of parody, with developer Neal Agarwal creating a free satirical game called "I'm Not a Robot." Users must solve a series of increasingly convoluted verification checks — scoring a point for each stage they pass, which eventually transcend into the absurd.

So, as machines get smarter, the answer may not be to find more difficult puzzles. "If a CAPTCHA can only be solved by someone with a Ph.D. in mathematics, then it's not very useful," Plesner said. "The internet needs to be used by everyone."

Can you match these ancient devices to their pictures? Find out with our computing quiz!

Network latency in bitcoin mining is driving massive energy waste ‪—‬ the annual equivalent of the total generation capacity of Switzerland's entire hydroelectric power system, scientists say. This wasted energy results from inefficiencies in the mining process and increasing competition among bitcoin miners.

In a new study published May 26 in the journal PNAS Nexus, the researchers aimed to provide a theoretical model to measure patterns within the networks powering bitcoin's distributed ledger system.

But they also calculated that in 2025, around 16,000 megawatts was wasted by fruitless bitcoin mining attempts, where competing mining efforts exert massive computational power to obtain the same units of bitcoin. This is roughly equivalent to the total generation capacity of Switzerland's 701 hydropower plants, according to statistics from the Swiss Federal Office of Energy.

It's important to note that this figure differs from the total energy consumed by bitcoin mining activity, estimated by researchers to stand at an annual level of 138 terawatt-hours, as of June 2024. This is higher than the annual energy consumption of several developed countries including Norway and the Netherlands.

Energy-guzzling crypto

Concerns around the environmental impact of bitcoin and other proof-of-work blockchain technologies have abounded in recent years.

In 2021, for example, bitcoin mining's water usage, primarily for liquid-cooled computer equipment, equated to more than the domestic water use of 300 million people in rural sub-Saharan Africa, according to a 2023 U.N. report.

Bitcoin is underpinned by a distributed ledger system, called a blockchain, which operates on a "proof-of-work" model. For a new unit of the digital currency to be generated, computing power must be used to solve a digital puzzle. In theory, the first entity to successfully "solve" the problem adds a new "block" of transactions to the ongoing chain and is granted a set quantity of bitcoin in return.

However, due to the explosion of interest in bitcoin as a financial trading asset, the competition for who can be the first to complete a block and claim the rewards has become incredibly fierce. A solution to the puzzle is based on computational power, with specialized hardware providing a greater advantage in speed. It has driven commercial entities to invest in building specialized data centers dedicated to such mining operations.

Because the race to be the first to mine a block is so competitive, the difference between first and second place can be just tiny fractions of a second. This often results in "accidental forks" — where two competing blocks are registered at almost exactly the same time.

In this scenario, the block with the longest chain of subsequent blocks built on top of it will eventually become a verified and legitimate part of the blockchain — earning its miners the bitcoin reward — while the competing block will be seen as invalid and worth nothing.

The energy needed to solve the proof of work and generate these "orphaned blocks" in the first place — as well as any subsequent blocks built on top of them before the winner is decided — is ultimately wasted.

"Despite their indication of a distributed network, accidental forks are an inefficiency of the Bitcoin protocol that leads to wasted computational resources (and thus energy), increasing the cost of network operations and its environmental impact to maintain a given level of security," the researchers wrote in the study.

According to the Crypto Carbon Ratings Institute (CCRI), a cryptocurrency analysis firm, bitcoin is the most dominant cryptocurrency by far, with a market capitalisation of more than $1.1 trillion — more than 80% larger than the next most popular currency, Ethereum. However, instead of proof-of-work, Ethereum uses a different form of consensus mechanism to establish block authorship, called "proof-of-stake," which is significantly less computationally intensive.

While other cryptocurrencies apart from bitcoin also use proof-of-work methods, bitcoin is around twice as large as its next nearest rival in this category, making it orders of magnitude more power-hungry.

Who rules the pool

Whereas previous models for analyzing fork rates treated all miners in the network as equal, this study considered elements such as network latency and geographic distribution, aiming to provide a "null model" — a baseline which can be used as a starting point to inform future analysis.

The model also allowed the researchers to quantify other notable trends, such as the distribution of "mining pools" — consortiums in which mining operators pool their efforts to maximize their potential success. They identified a decline in the dominance of Chinese mining pools from 2022 following the country's ban on bitcoin mining while also discovering high levels of consolidation at the upper level of the bitcoin mining industry.

The report found that just three mining pools produce over 50% of new bitcoin blocks. This is a problem because it risks a "51% attack," whereby unscrupulous miners enter fraudulent information into the blockchain by ensuring that they always produce the longest chain and, therefore, become validated.

This level of consolidation distorts the market for processing fees that bitcoin users pay to have their transactions included in the next block, the researchers added, and could thus allow miners to arbitrarily delay the inclusion of specific transactions.

A breakthrough in photonic chips could make large, costly, ultrafast lasers dramatically smaller, leading to portable and affordable imaging, diagnostic and information-processing devices, researchers say.

By using a decades-old overlooked laser architecture, scientists managed to fit an ultrafast laser onto a tiny photonic chip — a chip that uses light, rather than electricity, for computing operations.

In a new study published June 3 in the journal Nature, the team demonstrated that a tiny laser on the photonic chip could deliver 1.05 nanojoules of energy in 147-femtosecond (147 quadrillionths of a second) bursts — thereby competing with the output of laboratory-class ultrafast lasers.

Ultrafast lasers are used in a variety of applications, from precision manufacturing and eye surgery to biological imaging and atomic clocks, but the systems needed to power them tend to take up whole tabletops in labs or factories. Yet the powerful output of these laser pulses made them difficult to miniaturise onto photonic chips.

"For more than twenty years, a high-pulse-energy femtosecond laser on chip was widely regarded as a holy grail of integrated photonics," Tobias Kippenberg, a photonics professor at the Swiss Federal Institute of Technology(EPFL), said in a statement.

"Our result shows that it is not only possible, but that it can be achieved with a surprisingly elegant architecture that the integrated-photonics community had overlooked."

Forward-thinking breakthrough comes from looking back

Photonic chips manipulate light by using microscopic structures called waveguides — usually in the form of optical fibers or etched cavities — to carry information. They aren't particularly novel, and can be found in fiber-optic communications, medical sensors and lidar systems.

But photonic chips have previously struggled when handling high-powered, ultrafast lasers. That's because they need to contain light to extremely small waveguides, leading the light to interact strongly with itself and destabilizing the laser pulses.

To tackle this problem, the researchers looked at a laser architecture called the Mamyshev oscillator, created in 1998 by Pavel V. Mamyshev, a physicist and engineer at Bell Labs.

This oscillator, which has received little attention in the world of photonic chips, works by placing a nonlinear waveguide between two optical filters. This causes a high-intensity laser pulse to expand into a broader range of colors that can then pass through both filters while weaker light, which can cause laser destabilization, is blocked out. This technique essentially means that a high-intensity laser pulse can be maintained.

Because the Mamyshev oscillator doesn't require extra components to manufacture on a chip, it presents an attractive design for use on photonic chips. And although the laser cavity needed to direct an ultrafast laser is 16.5 inches (42 centimeters) long, it can be folded to occupy around the same area as a match head. This can't be done with conventional fiber-optic-based lasers, often used in photonic chips.

That takes care of the size, but the cost of ultrafast laser systems is another challenge. But because photonic chips can be fabricated using silicon wafers in the same fashion as computer chips, more than 1,000 laser cavities could potentially be produced in a single batch, the researchers said. As such, photonic chips with ultrafast laser capabilities could be produced at scale, in turn reducing manufacturing costs and even expanding their use.

Photonic chips capable of handling ultrafast lasers could, in the future, lead to portable tools for tasks like detecting pollutants or performing advanced medical diagnostics in the field, the researchers noted in the study. The technology also opens the door to smaller atomic clocks that can benefit navigation and future communications.

A Chinese system has become the world’s most powerful supercomputer, surpassing American machines for the first time since 2021.

LineShine, installed at China’s National Supercomputing Center in Shenzhen, clinched the top spot in the 67^th TOP500 ranking of the world’s most powerful supercomputers. The new system has already been used in a range of fields, giving developers another route to achieve supercomputing power.

The machine, which came online in the first half of 2026, can reach speeds of 2.198 exaFLOPS — where 1 exaFLOP is 1 quintillion (10¹⁸) floating-point operations, or mathematical calculations, per second (FLOPs) — making it the only supercomputer on the planet to exceed 2 exaFLOPS per second. It's also the first time China has hosted the world's fastest supercomputer since 2017.

A FLOP is a type of calculation used to benchmark computing performance. Exascale supercomputers can perform more than 1 quintillion of these operations every second. In comparison, home computers can perform roughly 5 trillion FLOPS.

According to TOP500, LineShine can achieve speeds about 22% faster than El Capitan, a supercomputer housed at Lawrence Livermore National Lab in California that had previously held the top spot since November 2024.

The system’s computing power is the result of "a comprehensive breakthrough in a series of core technological barriers," according to a translated statement from China's National Supercomputing Center.

Unlike many other supercomputers, LineShine uses only central processing units (CPUs) to perform calculations. Other systems rely on both CPUs and graphics processing units(GPUs), which run many jobs simultaneously by dividing tasks among smaller, specialized cores.

Since 2018, the U.S. government has restricted exports of semiconductor chips to China, including GPUs. However, startups such as DeepSeek have wrangled other technological advancements to train artificial intelligence (AI) models with fewer and less powerful GPUs than comparable systems such as ChatGPT.

LineShine "represents a historic leap forward for China's supercomputing field, breaking through foreign technological blockades and building an independent and controllable software and hardware system," the statement read.

The system has already been used on projects in multiple research areas, including atmospheric science, drug discovery and AI, according to the National Supercomputing Center. In general, supercomputers perform extremely complex calculations at speeds much faster than traditional computers can handle, allowing them to solve problems that would otherwise take too long or cost too much to address.

The fastest supercomputers utilize a range of different designs and processors, showing that high-performance computing doesn’t rely on any one single method.

"The list demonstrates that there is no single dominant technology path to leadership-class computing; instead, vendors are pursuing a variety of CPU, GPU, APU, and custom-accelerator approaches coupled with different interconnect and system designs," TOP500 representatives said in a statement.

Following LineShine and El Capitan, two supercomputers at U.S. national laboratories and one in Germany claimed spots three through five on the TOP500 list. Machines in Italy, Switzerland, Japan and the U.S. round out the top 10.

Can you match these ancient devices to their pictures? Find out with our computing quiz!

As I sip coffee in my Berlin apartment and fire a question at Google's AI chatbot Gemini, it's easy not to think about the energy it takes to generate a response. Once the signal reaches my router, it whizzes, I assume, through copper wires or fiber-optic cables to one of Google's data center hubs. Somewhere inside the data center's labyrinthine halls of stacked processors, my query gets converted into numbers and undergoes billions of computations to determine context and meaning. The answer, once assembled, races back, in the blink of an eye.

Data centers — the beating hearts of the internet, powering everything from email to web searches — have existed for decades, but with the growing popularity of AI to generate text, images and video, they're using more energy than ever. According to Google's own estimates, processing a median-length text prompt with its AI assistant Gemini consumes around 0.24 watt-hours.

These amounts, individually small — 0.24 watt-hours is equivalent to watching TV for about nine seconds — are adding up fast. In March 2026, OpenAI estimated that more than 900 million people use its AI chatbot, ChatGPT, every week, tallying billions of queries daily.

The exact amount of electricity consumed by data centers, globally or in the United States, which hosts more than any other nation, isn't publicly reported by all tech companies, says Eric Masanet of the University of California, Santa Barbara, who researches data center sustainability. But according to the most recent estimates by the International Energy Agency, US data centers guzzled some 224 terawatt-hours of electricity in 2025 — more than 5 percent of the country's electricity use. That's a significant uptick from an estimated 1.9 percent consumed in 2018, well before the mainstream surge of generative AI.

This electricity use seems set to soar. In the race to secure market leadership for generative AI products, companies like Google, Meta, Amazon, OpenAI, Anthropic, Microsoft and Oracle are investing tens to hundreds of billions of dollars to build AI-focused data centers. Compared to data centers of the pre-AI days that consume, say, 100 megawatts of electricity — enough to power 83,000 homes with average demand — the newcomers are often "hyperscale" and can use a gigawatt or more, or roughly a tenth of the electrical capacity of Los Angeles.

Masanet and other experts have been alarmed to see much of this demand met by plants powered by fossil fuels, such as gas, whose burning releases planet-warming carbon dioxide. A key reason is that data centers are often constructed in places without abundant renewable energy sources like hydropower, geothermal, solar or wind.

Tech companies often offset emissions by investing in renewable energy elsewhere. But unless those clean energy plants make more energy than the data centers use, this strategy — at best — keeps CO₂ emissions of centers in stasis rather than reducing them to a net of nothing, important for halting global warming. "For every megawatt for which we install fossil fuel power," Masanet says, "it sets us back on our progress."

And that's not considering the resources spent on manufacturing the hardware that fills new data centers, or the impacts on communities living near them, which often suffer from air and noise pollution from gas plants and possible strain on local water resources, which are used to cool the data centers.

Although forecasts for AI's energy impact remain devilishly tricky, especially since the size of payoffs from investments in AI are uncertain, it's clear to experts that energy-saving strategies are urgently needed. Without them, according to one 2025 estimate, US data centers could soon be releasing the equivalent of 24 to 44 megatons of CO₂ annually, the latter equivalent to the annual emissions of Norway.

And so computer scientists and engineers are rethinking some of the power-hungry hardware and software that fuel AI. They're working to develop energy-saving algorithms and processor designs, and carefully considering where, and how, data centers are constructed.

"AI's energy cost is not an accident: This is basically a product of how our systems are built," says Fengqi You, an expert in energy systems at Cornell University. But with the right mix of solutions, he says, "we could really reshape the trajectory."

The roots of AI's energy problem

To comprehend AI's energy cost, it helps to understand large language models (LLMs) — the lifeblood of AI text generation tools such as chatbots and AI assistants — specifically, ones based on a design described in 2017 by the machine-learning laboratory Google Brain. This design, transformer architecture, can process text at lightning speed by simultaneously taking each word and weighing its relationship to every other word it sees. It "learns" which words go together by computing how strongly each word relates to all other words in a text, examining each word in many contexts. (A similar design is used for AI image and video generators.)

On a computational level, this happens by converting words or word fragments into numbers and performing additions and multiplications between them. Key to the speed is being able to do these calculations in parallel, made possible by graphic processor units (GPUs) — mostly manufactured by the company NVIDIA — originally invented for rapid 3D rendering of imagery during gaming.

The initial training of an LLM, required to learn all these relationships, consumes vast amounts of energy. Because each word it trains on must be weighed against all others in a given chunk of text, the number of computations the model performs — hence the energy required — increases quadratically relative to the length of text (i.e., doubling the length of text quadruples the number of computations). That adds up quickly given that most LLMs are trained on massive swaths of publicly available internet text. Some estimates suggest that training GPT-4 — the iteration of ChatGPT that launched in 2023 — guzzled between 50 and 60 gigawatt-hours of electricity, enough to power San Francisco for three to four days.

But experts are more worried about the energy costs of using the models to generate data once they've been trained, a process called inference. "You train once, then you inference for a billion people in the world," says Mosharaf Chowdhury, an AI systems expert at the University of Michigan who has been measuring the electricity usage of a handful of large language models that have been made publicly available.

This process is surprisingly inefficient: Each time transformer models generate a word — by selecting the one with the highest probability of following the previous word, given context — they put the query and partially written answer through the model. In doing so, they apply all of the parameters they've calculated during training to understand language patterns — which number in the hundreds of billions or even trillions.

"The fact that you have to do a lot of calculations for a single word to be added — that’s a problematic thing," says Günter Klambauer, an AI expert at Johannes Kepler University in Austria.

Tweaking AI software to save energy

This recognition has triggered interest in smaller language models specialized to specific tasks. These are trained more narrowly, have fewer parameters — say, tens or hundreds of millions — and perform substantially less computation than larger models. In one 2025 paper published by UNESCO, computer scientist Ivana Drobnjak of University College London and colleagues compared energy consumption of Meta's language model Llama-3.1 with smaller AI models dedicated to particular tasks — ones called DistilBART and t5-small-xsum for summarization, and others for translation or answering questions. When used for their respective tasks, the smaller models consumed more than 90 percent less energy than Llama 3.1 on the same job.

And so computer scientists have been driven to build a similar kind of task specialization into LLMs themselves. In "mixture of expert" models, only particular parts of one big model are activated for certain tasks. These parts "learn to handle different patterns in language," Drobnjak says.

This is thought to be one reason why R1, an LLM developed by the Chinese company DeepSeek, reportedly consumed significantly less energy than other models (independent experts have raised doubts about those figures). Udit Gupta, an expert in electrical and computer engineering at Cornell Tech, says that LLMs like Gemini or ChatGPT are similarly routing queries to more specialized sub-models. "There's a lot of work being done on how to assess the complexity of the query or task that's coming from users and then find the right model," Gupta says. (While Google spokesperson Ralf Bremer notes that the 0.24 watt-hours currently spent on processing median-length Gemini prompts is already 33 times more efficient than it was back in 2024, some experts suspect that processing queries with an LLM still consumes more energy than an equivalent web search.)

Scientists are also exploring different kinds of LLMs, to break what Klambauer calls the "quadratic curse" of transformer models.

One alternative, called a long short-term memory (LSTM) model, gets around this alarming energy increase by temporarily storing a kind of summary of the prompt that was inputted by the user plus the text generated so far, akin to recalling important plot points instead of an entire movie. That way, it only has to process the summary, rather than all the words in the full text to date, every time it generates a new word. This prevents LSTM's energy costs from skyrocketing as it responds to a query — using about 50 percent less energy than transformer-type models to process texts of around 8,000 words in length, Klambauer says.

LSTM models were developed in the 1990s but were abandoned because transformers could be trained much faster. But Klambauer says that recent advances have improved the performance of LSTM, now called xLSTM. He's working with the Austrian startup NXAI to further develop and optimize xLSTM, "because we think it's worth it for energy efficiency," he says.

But major tech companies have invested so many years and resources into developing transformer-based models that switching to other models would be costly, says Wolfgang Maaß, an AI and business informatics researcher at the German Research Center for Artificial Intelligence. "We have to see whether this becomes as dominant, or whether it finds a niche in the whole market."

Computing with wafers and light

Though experts say the fastest energy savings will come from software tweaks, some are also taking aim at the energy-hungry processing chips that fuel AI computations. Engineers have made chips increasingly efficient over time by packing more computing capacity into individual processors — reducing the energy required to shuttle data between chips that are working together to perform AI computations. Engineers have done this by shrinking the size of transistors — microscopic electrical switches that process data — inside the chips.

But because engineers are reaching the physical limits of how small transistors can be, "we need to think of alternate ideas to improve the designs," says computer architect Ajay Joshi of the Boston University Photonics Center.

One strategy is to make the chips larger. Dinner-plate-sized "wafer-scale chips" can pack nearly 70 times as many transistors as a single, postage-stamp-sized GPU and consume 143 times less electricity for communication than comparable GPUs, says computer engineer Rakesh Kumar of the University of Illinois Urbana-Champaign. Commercially produced by the California company Cerebras, wafer-scale chips have drawbacks, including a greater risk of damage during manufacturing. But because of their energy-saving and other beneficial features, "they would be very attractive to many hyperscalers and AI companies," Kumar says.

Many tech companies have improved energy efficiency by fashioning their own processors that are tailor-made for AI computations — such as Amazon Web Service's Trainium2 chip or Google's Ironwood Tensor Processing Units — according to statements from those companies. As for NVIDIA, the company's head of sustainability Josh Parker says its AI-specialized GPUs have come a long way from the ones used for gaming and are now designed to run AI tasks as efficiently as possible; other innovations, such as making the interconnections between GPUs more efficient, have also helped. "Over the past eight years, NVIDIA GPUs have improved 45,000 [times] in energy efficiency for large language model workloads," he says.

Engineers are also exploring alternative computing methods. Conventional AI processors calculate by encoding numbers in a binary system of ones and zeros, which is achieved by turning transistors on and off (representing the number 5, for instance, requires four transistors to represent the code 0101). But transistors can do more than function as binary switches allowing electron flow or not; they can also work as analog dials and hold intermediate voltages representing different numbers. That requires fewer transistors, and less energy, for computations. "People have known for decades that doing certain things in analog … can be a lot more energy efficient," Kumar says.

For example, electrical engineer Paul Manea of the German research institute Forschungszentrum Jülich and colleagues are working to develop devices called "gain cells" that are full of transistors working this way. Importantly, gain cells can both store the data required to process a query, and compute the answer. That overcomes another big energy bottleneck of conventional computing systems, where memory storage and computation occur on separate pieces of hardware.

That's especially problematic for transformer-based LLMs, because each time they generate a word, they must shuttle the query and partially written answer from memory to a processor. Manea and colleagues estimate that gain cells in lieu of traditional GPUs can reduce the energy guzzled by one of the most energy-consuming parts of transformer-based LLMs by four orders of magnitude. But it will take more refining before they can be more widely used, Manea says.

The notion of devices that both store and compute information is a key idea of "neuromorphic" computing, an up-and-coming field of computer engineering inspired by the human brain, which consumes orders of magnitude less energy than computers. Another brain-inspired invention is chips that encode information not in continuous data streams but — like human nerve cells — in the timing of voltage "spikes" propagating through the system. Allowing components to rest until they're needed "could potentially translate to less energy," says Eleni Vasilaki, an expert in bioinspired machine learning at the University of Sheffield in England.

Maaß, for example, is part of a team that received roughly $5.8 million from the German government to test neuromorphic chips, among other strategies, to reduce the energy required for AI models. Some brain-inspired chips are already commercially available, but the technology is still far from being attractive for mainstream computing, says nanoelectronics expert Tony Kenyon of University College London, whose team recently received $17 million from the UK government to develop neuromorphic computing.

Other scientists are developing chips that process information not with electrons but through the interaction of photons — particles of light — with matter (fiber-optic cables, which encode and transmit data as light pulses, are used around the world). With photons, more information can be transmitted at the same time, and signals can be altered much faster, says Elena Goi, a photonic computing researcher at Friedrich Schiller University Jena in Germany.

Several companies have developed chips that can perform some AI computations with optical methods, says Joshi; he recently estimated that manufacturing optical chips could consume up to an order of magnitude less energy than conventional ones of the same size. Joshi hopes that, "in 10 years, we would have a practical solution that can be deployed pervasively across the data centers."

Reshaping AI's energy trajectory

Even without reinventing how computers work, much can be done to reduce AI's impact not just on energy but also on water resources used for cooling data centers. Importantly, tech companies should reconsider where they build those centers, says energy systems expert You. Right now, existing US ones are concentrated in northern Virginia, which has limited water resources and renewable energy capacity compared with the Midwest, for instance. You recently estimated that better siting — along with energy-efficient hardware and software — could reduce future carbon and water footprints of US data centers by 73 percent and 86 percent, respectively.

Masanet adds that tech companies already with data centers across the country could at least train their models in strategic places. "Some companies like Google have been doing this: They shift their loads to follow renewables," he says. They also should address the electricity and resources spent on manufacturing processors for new data centers, as well as electronic waste as outdated tech is replaced every few years, he adds.

Minimizing e-waste by using hardware for longer periods and recovering old electronics is one of Amazon's sustainability strategies, according to a statement to Knowable Magazine; so is designing data centers in energy- and water-saving ways and investing in a slew of renewable and nuclear energy projects. "We'll continue to implement solutions that benefit our customers and the communities we operate in," says Brandon Oyer, Amazon Web Services' head of energy and water in the Americas.

Meanwhile, a press representative at Microsoft points to a number of sustainability initiatives the company has taken, including new cooling technologies, renewable energy investments and waste reduction. Google spokesperson Ralf Bremer emphasized the company's goal of reaching net-zero emissions across its operations by 2030 and replenishing 120 percent of the fresh water consumed by its offices and data centers by 2030. An OpenAI representative points to a press release outlining efforts to minimize water use and plans for solar energy generation at one of its campuses. Anthropic, Meta and Oracle did not respond to requests for comment by deadline.

Though tech companies are taking sustainability into consideration, their main objective is to rapidly build out data center capacity, says computer engineer Benjamin Lee of the University of Pennsylvania. He predicts that, eventually, they'll need to step up efforts to improve energy efficiency to reduce costs. Governments should help to accelerate this shift, Masanet says. So far, he and his team have counted nearly 220 policies introduced to address data center sustainability at the US state level, 18 at the federal level, and more from other countries, though not all were ultimately adopted.

"It's clear that governments around the world are beginning to take action," he says. However, he adds, "we also see some state and local governments with proposed policies that mostly aim to incentivize and accelerate data center builds."

AI's energy cost will ultimately be a balancing act: Will it save more resources through its problem-solving abilities deployed toward everything from finding cancer cures to improving logistics, than it demands? But though building a more frugal, energy-saving AI is important, so is carefully considering where AI is needed, Kenyon says. Is the world truly a better place, for example, with nonhuman "AI agents" providing customer support?

"I think it’s a common mistake, when a new technology comes in, to suddenly think, 'Well, everything has to adopt that new technology,'" he says. "That approach really isn't doing us any favors."

This article originally appeared in Knowable Magazine, a nonprofit publication dedicated to making scientific knowledge accessible to all. Sign up for Knowable Magazine's newsletter.

A photograph of Earth glowing in deep space, the moon's cratered horizon stretching across its foreground, caught many people's eyes in April 2026. Astronauts captured the image while aboard NASA's Artemis II mission, and like the famous Apollo 8 "Earthrise" image, the picture felt instantly real and inspiring for many.

But when almost anyone can fabricate a visually similar image in seconds from a text prompt using artificial intelligence, how do people decide which image is real?

The proliferation of AI-generated science images in public spaces is not simply a misinformation problem. As a researcher who studies visual science communication and public trust, I believe it also contributes to a crisis of trust in science in the age of AI, and the tools scientists have long relied on to establish visual credibility are losing their grip.

AI-generated images infiltrate science

AI tools are already changing how scientific visuals are created, shared and publicized.

Researchers use them to generate illustrations, create synthetic data, edit lab images and produce materials for education and public outreach.

While AI can help scientists communicate complicated ideas more creatively and efficiently, these same tools blur the lines between illustration, enhancement and fabrication.

In 2024, two papers were retracted after publishing AI-generated figures posessing biologically impossible structures. In April 2026, the New England Journal of Medicine retracted a paper after discovering that a clinical image had been manipulated with AI. These are just cases that came to mass public attention and are likely just the tip of the iceberg. Researchers have warned that AI-generated visuals pose growing threats in fields that depend heavily on visual evidence, such as materials science.

Academic publishers are beginning to adopt AI-detection tools. However, systems designed to detect fake images will almost always lag behind systems designed to create them. Many detectors can identify only image patterns they were trained to recognize. As new AI models emerge, developers must constantly obtain new data and retrain detectors to catch up.

The biggest concern are realistic-looking visuals that subtly distort scientific details while remaining believable enough to pass initial review.

Trust in scientific images

For decades, scientific images carried authority partly because they were difficult to produce. Creating microscope images, climate graphs and space photographs required expensive equipment, institutional resources and specialized expertise. Most people assumed such images represented true observations because very few people could make them.

Research in science communication, including my own, suggests that people judge scientific visuals using a few mental shortcuts. Does the image look technically sophisticated? Does it come from a trusted institution? Does it match what I already believe? Generative AI is undermining all three of these heuristics, or mental shortcuts.

Today, anyone can create a polished, scientific-looking image from a text prompt. Images are also detached from their original source when circulating online. When visual quality and institutional attribution become unreliable cues for judging the credibility of science images, people tend to fall back on something else: their own prior beliefs.

As a result, authentic scientific images that challenge someone's existing beliefs can now be dismissed as AI-generated, whereas fabricated images that confirm them are easily accepted as evidence. AI, in this way, may amplify motivated reasoning — that is, people's tendency to accept what they already agree with and question what they do not.

This shift matters because visuals have long served as evidence for scientific claims. Nonexpert audiences rely on images not only to see what scientists have discovered but also to develop an emotional connection and perceive credibility in the science being presented.

If audiences stop trusting visual evidence altogether, science loses one of its most powerful tools for public communication.

Transparency, not restriction

AI tools offer real benefits for researchers communicating their work to diverse audiences. The challenge is using these tools without quietly transferring AI's credibility deficit onto the science the images are meant to convey.

One practical path forward is for researchers to treat image provenance — where an image came from and how it was created — with the same seriousness they already apply to data provenance.

Scientists routinely disclose funding resources, study methodologies and conflicts of interest. Similar standards may now be necessary for scientific images. Was AI used to generate or modify this image? Is it a direct observation, a simulation or an illustration? What exactly does the image represent, and how was it verified? Can it be replicated by other researchers?

My colleagues and I found that people's familiarity with AI significantly shapes how they judge the credibility of AI-generated visuals. Those familiar with AI tools were more likely to view AI disclosure as a sign of transparency, and some rated clearly labeled AI-generated content as more credible than unlabeled content.

Transparency gives audiences the necessary context to evaluate what they are seeing, but it may not resolve every dispute about how images are made. Responsible use of AI-generated scientific images will require honesty, adherence to professional norms and the collective development of evidence-based standards across fields.

Why authentic images remain powerful

The original Apollo 8 "Earthrise" photograph of 1968 carries significant emotional impact. So do the Artemis II images of 2026.

What makes them meaningful is not simply their beauty. It is their traceable connection to scientific reality. When people look at these photographs of planets, they also know there are astronauts, physical cameras, documented missions and verifiable observations behind the images. In this sense, authenticity is a documented relationship between an image and the world.

In the age of generative AI, scientific institutions can no longer assume audiences will automatically trust their visuals. Trust now depends on transparency, documentation and clear communication about how visual evidence is produced.

Without guidelines and standards, science risks entering a world where every image can be questioned and no image carries inherent credibility.

This edited article is republished from The Conversation under a Creative Commons license. Read the original article.

Who is responsible for AI's output? Artificial intelligence (AI) companies like OpenAI maintain that they are not. In fact, their terms and conditions in 2023 stated that responsibility lies solely with the user. A German court disagrees.

On June 9, a Munich court (subject to appeal) ruled that Google can be liable for false claims produced by its AI summaries, drawing a sharp line between ordinary search results and machine-generated assertions. In other words, AI companies must be held legally responsible for the output that is created by their systems and pushed to users.

The court's logic was simple but profound: Search results point outward to sources, while AI summaries speak in Google's own voice. That distinction matters because it goes to the heart of what kind of speech deserves protection — and what kind is subject to legal scrutiny. The U.S. should follow the German court's lead. In the absence of such provisions, the entire burden of discerning truth from falsehood falls on the reader.

In the U.S., the First Amendment is intended to protect the right to speak, argue, persuade and offend. But freedom of speech is not free of caveats. It does not allow people to incite others to commit crimes, to threaten or to defame, for example. And if speech causes material harm, speakers can be held liable for those harms. When a company chooses to put a synthetic answer engine between users and the web, it is no longer merely hosting speech; it is producing an amalgamation of complex mathematical expressions that, outputted as text, resemble human speech. AI companies want this text to enjoy the same protections user-generated text has, while simultaneously dodging all the responsibility associated with being a speaker.

The roots of this dilemma go back to the 1990s, when the advent of online forums and social media created a new problem. Unlike traditional publishers, forum hosts needed to provide a platform for their users' voices, without being liable for what their users were saying. This problem was addressed with Section 230 of the Communications Decency Act, enacted in 1996. Section 230 was a bipartisan amendment written to preserve the internet as a space where ordinary people could speak (or post) without the forum host becoming liable for every third-party post.

That broad immunity reflected a democratic judgment: If the law made platforms responsible for all user content, many would censor aggressively or stop hosting speech altogether. This would limit free speech. Section 230 was meant to protect the ecosystem of human expression. In this sense, hosts of online spaces can be seen as providing a public square where speech occurs.

The lawmakers who passed Section 230 three decades ago could not have foreseen a world populated by chatbot-generated text. As such text increasingly leads to real-world harms, lawsuits are proliferating and tech companies are deploying a number of often-contradictory legal strategies to avoid culpability. In some cases, they are arguing that AI-generated text is not speech, but rather simply a tool, and that companies are therefore protected as "carriers," not "publishers" by Section 230's protection of a public forum for free expression.

But the companies deploy this argument only when it suits them.

In other cases, they are increasingly reaching for free-speech language to defend AI-generated text because free-speech protections provide broad legal immunity. For example, in a Florida wrongful-death lawsuit against Open AI (maker of ChatGPT), a plaintiff has alleged that the company’s chatbot pushed a 14-year-old to take his own life. OpenAI argued that the chatbot was protected by the First Amendment, though the judge dismissed that defense and allowed the case to proceed.

Neither of these arguments is convincing. AI companies are not merely providers of a public forum, as the words produced by their AI summaries and chatbots are generated by the company's products.

Similarly dubious is the claim that bots should be seen as equal participants in a public square. This is a category error. Free speech is a human right — it protects people as speakers and listeners in a democratic public sphere. Bots do not vote, deliberate, dissent, worship or participate in civic life. They generate text, but they do not possess a moral and political standing. Bots have no skin in the game.

What, then, justifies constitutional protection in the first place? Extending the strongest speech protections to machines would not defend liberty; it would confuse "botput" with free expression. It would, in actuality, extend the strongest free-speech protection to companies. But that requires a separate line of argumentation that ought to be agreed upon by society.

The Munich court's limited and nuanced way of governing "botput" provides a clear way forward.

Given its history with Nazism, Germany does not enshrine free speech quite the way the U.S. does. But the German court's arguments still provide a useful template for a future U.S. ruling.

The Munich court held that if a system simply points users to sources, it resembles traditional search and should continue to enjoy broad protection afforded to aggregators. If it synthesizes claims, imitates the tone of authority, and offers a single authoritative answer generated by an AI, it should carry corresponding duties of care that entail liability for the company.

The need for such safeguards is only growing. AI-generated summaries can be copied instantly, scaled globally, and repeated across interfaces until a falsehood becomes regarded as "truth." That is not a hypothetical concern; it is already happening.

Moreover, it is important to remember that the original intention of Section 230 was to insulate platforms from liability for third-party posts, not their own text.

This is not an anti-innovation argument. AI can be helpful, efficient and genuinely transformative. The law should encourage useful tools while insisting that the companies deploying them remain responsible for the foreseeable harms of their products.

We need clearer rules that keep the internet free for people while preventing machines from laundering falsehood into authority. The German ruling points toward that future. The sooner U.S. law and policy follow, the better chance we have of preserving our shared reality and a healthy democracy.

Opinion on Live Science gives you insight on the most important issues in science that affect you and the world around you today, written by experts and leading scientists in their field.

Researchers have created a new chip that turns one of quantum computing's biggest frailties into a programmable feature. They say this first-of-its-kind experiment could carry implications for developing error-corrected, fault-tolerant quantum computers in the future.

Unlike digital bits in a classical computer, which are represented as either "on" or "off," a quantum bit (qubit) has a much higher failure rate — roughly 1 in 1,000, compared with 1 in 1 billion for digital bits. That's because quantum computers are susceptible to "noise" — interference that's often cited as the biggest barrier preventing quantum computers from being more capable than the fastest supercomputers.

As engineers develop quantum systems that are large enough in scale to perform useful functions, the amount of noise generally increases. Scientists can combat this noise using various error-correction techniques. But despite recent progress in this field, the challenge of developing a truly fault-tolerant quantum computer remains.