As AI becomes more integrated into everyday life and more sophisticated, authorities face new challenges in identifying and responding to bad actors — particularly in cases of swatting.
Swatting involves making false emergency reports to 911 or local law enforcement, often claiming there is an ongoing bomb threat, active shooter, or another life-threatening event that could trigger an extensive police response, like a SWAT team, NPR reports. The calls divert public safety resources from actual emergencies and are often intended to harass, intimidate, or retaliate against targeted individuals, including public officials, the outlet notes.
While swatting calls are not new to 911 dispatchers, incidents targeting several schools nationwide over the summer highlight an emerging challenge — bad actors using AI-generated voices to make their hoax reports sound increasingly convincing.
“Those voices [of swatting callers] used to be pretty easy to recognize, but now they’re natural sounding. They can insert pauses. They can use words like, ‘uh’, or ‘maybe, um, let me think about it.’ And it sounds very natural,” said Mel Maier, the CEO and executive director of the Association of Public-Safety Communications Officials (APCO).
The Growing Threat Of Swatting Calls
A June report from the APCO and the Public Safety Threat Alliance, a cybersecurity information hub from Motorola Solutions, examined the prevalence of swatting calls targeting 911 centers. The report, based on 766 responses from public safety communications professionals, found that approximately 14% of dispatchers had been targeted by swatting attempts involving AI tools, per NPR.
Brandon May, an assistant professor of forensic psychology at the Florida Institute of Technology, said inexpensive software now lets people alter voices and create AI-generated images and videos that can be hard to distinguish from authentic recordings. He notes that increasingly sophisticated AI tools, including chatbots, have even been used to deceive people into giving away their life savings.
“You couple that in a high-stakes, high-consequence 60- to 90-second call — it is going to be nearly impossible to detect,” May told NPR in another report.
With AI-assisted hoaxes on the rise, APCO and the National Emergency Number Association (NENA) are developing training to help public safety communications professionals recognize and respond to these emerging threats.
April Heinze, NENA’s vice president and chief of 911 operations, outlined some signs of a potential swatting call, including callers who avoid dialing 911 directly. Doing so can provide dispatchers with identifying information, including the caller’s location, callback number, and phone subscriber details, NPR notes. Instead calls may come in to police departments or court houses and get routed to 911. Still, she emphasized that emergency responders will always treat reported threats seriously and respond as if they are real.
“Even if there are indicators that it is a swatting incident. They are always reporting a life-or-death situation, and if it’s a life-or-death situation, you cannot ever take chances on that,” Heinze said.
