AI Security System Flags Snack Bag, Leads to Student Detainment at Maryland High School

Baltimore County, Maryland – A student at Kenwood High School was reportedly handcuffed and subjected to a search after an artificial intelligence (AI) security system misidentified a bag of chips as a potential firearm. The incident, reported on October 25, 2025, at the school in Baltimore County, has drawn attention to the operational parameters and potential for false positives in AI-driven security deployments within educational environments. Taki Allen, the student involved, provided an account to CNN affiliate WBAL. Allen stated that he was holding a Doritos bag with both hands and one finger extended when the AI system issued its alert. Recounting the immediate aftermath, Allen stated, \"They made me get on my knees, put my hands behind my back, and cuffed me,\" following the system's erroneous detection. Kenwood High School Principal Katie Smith addressed the school community regarding the event in a statement shared with parents. Smith confirmed that while the school’s security department had reviewed and subsequently canceled the initial gun detection alert, she was reportedly unaware of this cancellation at the time. Consequently, Smith escalated the situation by notifying the school resource officer, who then contacted local police authorities, leading to the student's detainment. The sequence of events highlights potential communication gaps in the deployment of advanced security technologies. Omnilert, the company responsible for operating the AI gun detection system deployed at Kenwood High School, acknowledged the incident and its outcome. In a statement provided to CNN, Omnilert conveyed, \"We regret that this incident occurred and wish to convey our concern to the student and the wider community affected by the events that followed.\" Despite the system's erroneous identification of the snack bag as a weapon, the company maintained that \"the process functioned as intended,\" indicating that the system adhered to its programmed protocols for flagging potential threats, regardless of the ultimate false positive. The event underscores ongoing discussions regarding the accuracy, reliability, and ethical implementation of AI security solutions, particularly in sensitive settings such as schools. AI-powered surveillance and detection systems are increasingly being adopted across public and private sectors for various security applications. This incident highlights specific challenges associated with minimizing false positives while ensuring effective threat detection and maintaining appropriate response protocols. Further details regarding the school's internal review of the incident and any potential adjustments to security protocols were not immediately available following the report.