AI Misidentifies Doritos Bag as Gun, Police Respond to Teen

An incident at Kenwood High School in Baltimore County has raised serious concerns about the reliability of AI-powered gun detection systems. A 16-year-old student, Taki Allen, was wrongfully detained after an AI misidentified a bag of Doritos chips as a firearm, leading to a significant police response.

On March 15, 2024, Allen was enjoying his snack outside the school following football practice when he suddenly found himself surrounded by multiple police vehicles. “It was like eight cop cars that came pulling up for us,” he recounted to WBAL-TV 11 News. The situation escalated quickly as officers approached him with weapons drawn, instructing him to get on the ground. “I was like, ‘What?’” Allen said, recalling the moment he was ordered to kneel and was subsequently handcuffed.

The officers conducted a thorough search and soon realized that Allen was unarmed. “I was just holding a Doritos bag — it was two hands and one finger out, and they said it looked like a gun,” he explained. This alarming incident highlights the significant flaws in current gun detection technologies, which are increasingly being implemented in schools across the United States.

Concerns Over AI Technology in Schools

The incident has sparked conversations around privacy and the potential for AI systems to misinterpret innocent objects as threats. Gun identification software, despite its intended purpose, has not effectively prevented actual school shootings, such as the recent tragedy at Antioch High School in suburban Nashville.

Additionally, there are concerns regarding racial bias within these systems. Critics argue that students of color may face disproportionate scrutiny, as seen in Allen’s case. Omnilert, a Virginia-based startup responsible for the technology used at Kenwood High School, has been criticized for its approach. The system analyzes footage from over 7,000 school cameras for suspicious activity and alerts law enforcement in real time.

According to a spokesperson for Omnilert, Blake Mitchell, the technology mistakenly verified the Doritos bag as a weapon because the image resembled a gun. He acknowledged that even upon review, the bag could still appear to be a firearm to some observers. “Even as we look at it now, with full awareness that it’s not a gun, it still looks like to most people like one,” Mitchell stated.

Despite calling the incident a “false positive,” Omnilert emphasized that the system functioned as intended by prioritizing safety through rapid verification. This incident raises critical questions about the appropriateness of such technologies in educational settings and whether they genuinely contribute to student safety.

The Impact on Students and Community Response

For Allen, the experience has left a lasting impact. He expressed fear about returning to school, stating, “If I eat another bag of chips or drink something, I feel like they’re going to come again.” His concerns are echoed by his family, particularly his grandfather, Lamont Davis, who criticized the excessive police response. “There was no threat for eight guns to be pointed at a 16-year-old,” Davis told the Baltimore Banner.

Moreover, Allen reported that he has not received an apology from the school administration. “They just told me it was protocol,” he said, adding that he expected at least some acknowledgment of the distressing situation.

The incident at Kenwood High School serves as a stark reminder of the potential consequences of relying on flawed technology in environments meant to nurture and protect students. As the debate around gun control and school safety continues, the role of AI in maintaining security remains under scrutiny.

This troubling event underscores the need for greater oversight and a reassessment of how technology is employed in educational settings. As schools increasingly adopt these systems, it is imperative to ensure that they do not compromise the safety and well-being of students, particularly those from vulnerable backgrounds.