ShotSpotter AI Gunshot Detection System Led to Wrongful Police Raids and Community Harm

ShotSpotter, an acoustic gunshot detection system deployed in over 100 US cities, uses AI algorithms to analyze audio captured by networks of microphones to identify potential gunshots and alert police. The system was marketed as providing rapid response capabilities to gun violence, with contracts worth millions of dollars in cities like Chicago, New York, and San Francisco. Investigative reporting and academic studies beginning in 2021 revealed severe accuracy problems with the ShotSpotter system. The MacArthur Justice Center found that in Chicago, 89% of ShotSpotter alerts in 2019-2020 resulted in police finding no evidence of gunfire. A separate study by the City of Chicago Inspector General found that only 9% of ShotSpotter alerts led to evidence of a gun crime. The system frequently misidentified fireworks, car backfires, construction sounds, and other loud noises as gunshots. The deployment pattern of ShotSpotter sensors disproportionately covered predominantly Black and Latino neighborhoods, leading to discriminatory policing impacts. These false alerts triggered aggressive police responses including SWAT raids, with officers arriving expecting active shooters. In several documented cases, innocent residents were subjected to forced entries, property damage, and traumatic encounters with heavily armed police responding to non-existent gunfire. Chicago terminated its $33 million ShotSpotter contract in 2024 following years of criticism and community advocacy. Other cities including San Antonio and Charlotte also ended their contracts after similar accuracy and bias concerns. Multiple civil rights lawsuits have been filed alleging that ShotSpotter enabled discriminatory policing practices that violated constitutional rights of residents in targeted neighborhoods. The incident highlighted broader issues with deploying AI systems in law enforcement without adequate testing for bias and accuracy. ShotSpotter's proprietary algorithms were not subject to independent auditing, and the company reportedly modified audio classifications after police found no evidence of gunfire, raising questions about data integrity and feedback loops that could mask poor performance.