The defendant's searches on the AI assistant were presented as evidence in an arson case that resulted in one of the deadliest fires in Los Angeles history.
U.S. prosecutors have used a suspect's ChatGPT search history as evidence in an arson trial. The case involves Jonathan Rinderknecht, accused of starting a fire on January 1, 2025. The incident triggered one of the deadliest fires in Los Angeles history.
To support the charges, the prosecution cross-referenced different types of digital and testimonial evidence. The body of evidence included geolocation data extracted from the defendant's iPhone, security camera footage, and witness testimony. The addition of AI interaction records marked an expansion in the scope of digital evidence used in criminal proceedings.
According to The Verge, prosecutors presented the ChatGPT logs to demonstrate the defendant's state of mind and intentions before and during the spread of the fire. The use of interactions with AI assistants as a criminal investigation tool is a recent practice, but one that is likely to gain traction as these technologies become more popular among users.
The defense, in turn, questioned the validity and context of these searches on the AI platform, resulting in requests for a mistrial. Lawyers argue that interactions with AI tools can be misinterpreted or taken out of context, raising legal debates about the limits of using algorithm-generated data in court.
The case highlights the emerging legal challenges in the wake of the massive adoption of artificial intelligence. Just as browsing histories and private messages are already routine in forensic investigations, queries to language models are now coming under the scrutiny of authorities, opening a new chapter in the intersection between technology and the justice system.
Prosecutors presented the defendant's ChatGPT logs alongside geolocation data, security footage, and witness testimony to demonstrate his state of mind and intentions before and during the Los Angeles fire.
The defense requested a mistrial, arguing that interactions with AI tools can be misinterpreted or taken out of context, which raises legal debates about the limits of using algorithm-generated data in court.
It marks an expansion in the scope of digital evidence. Just as browsing histories and private messages are routinely used in forensics, queries to language models are now coming under legal scrutiny as AI adoption grows.