AI-Generated False Testimony: Attorney Fined and Held in Contempt by New Mexico Supreme Court
A defense attorney appealing a murder conviction was fined and held in contempt by the New Mexico Supreme Court for submitting a brief that contained false police testimony and fabricated witnesses created by OpenAI's ChatGPT. The attorney, Stephen Aarons, admitted to using the AI program to help prepare the court filing but failed to verify its accuracy. The court found Aarons lacked remorse and fined him $5,000, referring him to an attorney disciplinary board for investigation.
Aarons had utilized ChatGPT to summarize trial proceedings for the defendant's appeal but underestimated the AI's ability to create false information. He expressed remorse for the mistake and hoped the disciplinary board would consider it an honest error. The court's sanction is part of a trend where lawyers have faced disciplinary action for submitting court documents generated by AI tools without proper verification.
The fabricated material in Aarons' filing included fictional witness testimony in a criminal appeal for Oscar Renee Sandoval, who was convicted of murdering the mother of his children. The appeal is ongoing, now assigned to a public defender, Kim Chavez Cook. The district attorney's office for Doña Ana County and OpenAI did not provide comments on the matter. Aarons explained to the court that he fed case materials to ChatGPT, expecting a reliable summary but ended up with inaccurate information.
In conclusion, the New Mexico Supreme Court's decision to fine and hold attorney Stephen Aarons in contempt for submitting a court filing with false information generated by an AI program highlights the importance of verifying AI-generated content in legal proceedings. The case serves as a cautionary tale for legal professionals relying on technology and underscores the need for diligence in ensuring the accuracy of court documents.