US Lawyer Blames ChatGPT for Citing Fake Witnesses in Murder Case

A New Mexico defense attorney, Stephen Aarons, has been fined $5,000 and held in contempt by the state's Supreme Court after submitting a murder appeal brief that included fabricated police testimony and non-existent witnesses, which he attributed to using OpenAI's ChatGPT. The court determined that Aarons failed to verify the information generated by the artificial intelligence program, which also misrepresented actual testimony and legal precedents in the appeal for his client, Oscar Renee Sandoval.

Aarons stated he used ChatGPT to summarize trial proceedings, claiming he was unaware of the extent to which AI could "hallucinate" facts. As a result of his actions, the New Mexico Supreme Court struck down all briefs he submitted for the appeal and appointed new counsel for Sandoval. Aarons has also been referred to an attorney disciplinary board for further investigation into his conduct.

This incident marks another instance in a growing number of cases where legal professionals have faced sanctions for relying on unverified, AI-generated content in court filings. Courts across the U.S. continue to emphasize the critical need for attorneys to exercise due diligence and confirm the accuracy of all information presented in legal documents, regardless of how it was generated.