Almost everyone knows what an Amber Alert is, and most of us have seen of heard at least one: A short description sent to road signs, mobile devices, and media stations of a recently missing child, along with information about where the child was taken from, the possible abductor's vehicle, and so on.

In theory, it seems like a great idea, allowing the public at large to help police find a missing child in the most important few minutes after an abduction. The problem is that most Amber Alerts are false alarms that do more harm than good.

Here is a sample of Amber Alert cases issued in the past two weeks:

  • In Oshkosh, Wisconsin, police issued an Amber Alert on Aug. 19 after a young boy reported seeing three men force a young girl into a car. Yet no child was reported missing, and the boy's report is suspected to be a prank or misunderstanding.
  • On Aug. 21, an Amber Alert was issued in the search for a 7-year-old Boise, Idaho, boy abducted by his mother and an accomplice; an arrest warrant was issued for the mother, who had been estranged from the boy's father.
  • An Amber Alert issued for 8-year-old boy in Wilkinsburg, Pennsylvania, was cancelled on Aug. 31 after he was found safe with his stepfather.

Noticeably absent from this list is the very type of abduction that led to the creation of the Amber Alert system in the first place: stranger abductions. In each of these cases — and in many more — the police violated protocol and misused Amber Alerts.

Though many people assume that Amber Alerts are issued for any suspected abduction, there are strict criteria that police are asked to follow in deciding whether or not to send out a notification. First, police must confirm that an abduction has in fact taken place (and the child is not, for example, simply late coming home from school); the child must be 17 or younger; the child must be at risk of injury or death; and there must be enough information on the child and/or suspected abductor to be usable. Cases that do not fit all of these criteria — established by the Dept. of Justice — should not trigger an Amber Alert.

Statistically, abductions by strangers are very rare (though the exceptions—like the recent case of Jaycee Dugard — create the biggest headlines). The vast majority of child abductions (well over 90 percent) are committed by a non-custodial parent or relative and thus will not qualify for an Amber Alert. It is misleading to assume that a child who is "missing" is likely in imminent danger. Often a "kidnapped" or "missing" child is safe and sound with one of the parents or a family member.

One study examined the outcomes of the 233 Amber Alerts issued nationwide in 2004. The majority (70 percent) were false alarms that should not have been issued because they failed to meet the established criteria. Abduction hoaxes are another serious problem; there are several hoaxes and faked reports of abductions every week—often reported by children as a prank.

This high false-positive rate highlights a little-recognized inherent contradiction in the premise behind Amber Alerts. The idea is that issuing an immediate alert will help save lives because tens of thousands of ordinary citizens can become the eyes and ears of police when every second counts.

But in order for the alerts to be effective, the police must first determine that the suspected abduction is real, and the child is actually in danger. The reality is that it could take hours to establish the facts of the case, to be sure that the child isn't simply lost, or with a non-custodial parent or relative. Police are put in a very difficult position, under pressure from concerned parents and the public to issue an Amber Alert immediately in the hopes of saving a life, while not triggering a false alarm.

Because of the very high false alarm rate, many experts are concerned that the Amber Alert system is being overused. The more alerts are issued (and the more that turn out to be false alarms), the less effective they become. Just as car alarms have become so common in many urban areas that they are often ignored — triggered by anything from a loud noise to a cat — the fear is that Amber Alerts will also be ignored. The system only works if the threat is real.

In some states the Amber Alert system has been used as a template for other alerts, for example to notify the public of senior citizens (such as those with Alzheimer's or dementia) who may wander off and come to harm. Indiana, for example, has such a "Silver Alert" system. Though the intent is surely good, it raises the question of where we draw the line. If we extend the system from children to seniors, why not 18-year-olds, or adults, or even pets? Why not robberies, or stolen cars, or any crime that the police could use the public's help with?

There is no doubt that the Amber Alert system and its variations have helped recover some missing people. But with the majority of alerts being false alarms issued by police erring on the side of caution, the Amber Alert system is likely to become a victim of its own success.

Benjamin Radford is managing editor of the Skeptical Inquirer science magazine. He wrote about Amber Alerts in his book Media Mythmakers: How Journalists, Advertisers, and Activists Mislead Us. His books, films, and other projects can be found on his website. His Bad Science column appears regularly on LiveScience.