AI turns bodycam footage into text. Can it keep the speakers straight?
Turning people into frogs—and back into princes—is the stuff of fairy tales. But back in January 2026, this fate befell a Heber City, Utah, police officer when AI accidentally cast the spell. Transcription software picked up the dialogue from "The Princess and the Frog" playing at the scene where the police encounter was unfolding. When AI software used the recording to draft the report, it…
A recent study published in the Journal of Experimental Criminology highlights the challenges of using AI to transcribe body camera footage. Researchers found that while AI can accurately capture most of the text, it struggles to keep track of the speakers when they switch. This can lead to mistakes, such as attributing statements to the wrong person or losing parts of the conversation.
For instance, in one tense exchange between an officer and a civilian, AI mistakenly attributed three different statements to the same person. The researchers recommend that human review remains essential to ensure accuracy, especially when it comes to identifying speakers and preserving the context of the encounter. While AI can provide an independent log of events, it is not a substitute for human judgment.
Investigators and researchers must still check the transcripts to catch any missing or inaccurate information.
Written by urgent.news from Phys.org's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
