It triggers strange things in the brain when someone sees a mistake or a lie in a video summary.

Research from Georgetown and Washington universities reveals that AI-generated summaries can distort human memory, leading to false recollections. The study analyzed summaries from ChatGPT and Gemini, finding significant inaccuracies and omissions. In an experiment, participants who read misleading AI summaries had a much lower accuracy in recalling details from a video compared to those who read accurate summaries. The findings raise concerns about the reliability of AI in critical situations, such as police investigations. Future research will focus on real-world examples, particularly police body camera footage and AI-generated reports, to understand their impact on public perception and memory.