Swatters are using AI voices to deceive 911 dispatchers, and it's getting worse (text.npr.org)

In a disturbing trend, swatters are increasingly using AI-generated voices to impersonate real individuals in hoaxes aimed at 911 dispatchers, complicating efforts to manage emergency responses. A recent incident involved a bomb threat called in with an AI voice at a California school, highlighting how these synthetic voices can now mimic human speech so convincingly—complete with natural pauses and inflections—that they pose a significant challenge to emergency responders. According to the Association of Public Safety Communication Officials (APCO), approximately 14% of dispatchers reported dealing with calls using AI tools, a figure projected to rise as technology becomes more accessible. The implications for the AI/ML community are profound. Voice synthesis technologies, like those from ElevenLabs, allow users to create convincing fake identities at a fraction of the cost, making it easier for malicious actors to exploit the technology for swatting or other fraudulent activities. Emergency organizations such as APCO and the National Emergency Number Association (NENA) are now focusing on developing training for dispatchers to help them discern potential swatting calls. This includes creating detailed interrogation techniques that AI cannot easily replicate. Despite the challenges, dispatchers are committed to treating all life-threatening scenarios as real, ensuring public safety resources are deployed regardless of potential deception.