Jonathan Rinderknecht, 29, repeatedly asked ChatGPT to generate images of a city engulfed in flames and floods before he allegedly set a brush fire that became the massive Palisades Fire in Los Angeles on January 1, 2025 [1, 2]. Prosecutors say the fire destroyed affluent neighborhoods in Pacific Palisades and was motivated by Rinderknecht's anger at social and financial inequality [2].
During his trial in June 2026, jurors heard that Rinderknecht cursed at ChatGPT when it refused to produce violent images due to content policy restrictions. At one point he told the AI, "No, what are you f**cking doing? Stick to what you said initially. On the left a scene showing environmental destruction," according to court records [2].
Federal agents recovered data from two phones belonging to Rinderknecht, including chat transcripts with ChatGPT, which showed his repeated requests for the violent imagery [1, 2]. Despite his attorney's claims that Rinderknecht is innocent and a scapegoat for firefighters' failure to keep the fire from reigniting, prosecutors argue he intentionally started the blaze [2].
US District Judge Anne Hwang barred the government from presenting the AI-generated images to the jury but allowed the prosecutors to show copies of the prompts Rinderknecht used in his requests [2]. Rinderknecht himself said during proceedings, "That's the reality," acknowledging the situation in court [2].
The trial remains ongoing with no verdict yet reached.
