Taking AI too far? New Mexico Supreme Court fines lawyer $5,000 for presenting ‘fake testimony’, ChatGPT-generated brief
New Mexico Supreme Court fined a defence lawyer $5,000 for submitting a brief containing ‘fake police testimony’ and ‘fabricated witness statements’ generated by OpenAI's ChatGPT.
The New Mexico Supreme Court has imposed a $5,000 fine on defence lawyer Stephen Aarons for submitting a brief containing fabricated police testimony and witness statements generated by OpenAI's ChatGPT. Aarons, representing Oscar Renee Sandoval in his appeal against a murder conviction, was found in contempt for not verifying the accuracy of the filing.
Sandoval was convicted of killing his children's mother and sentenced to life in prison last year. During a hearing on August 21st, Aarons explained that he had used ChatGPT to summarize trial proceedings, but did not realize the extent to which the AI could generate false information. The court expressed disbelief at Aarons' unawareness of AI's potential for mistakes, with Justice C Shannon Bacon stating that relying on AI hallucinations is a daily headline issue.
Aarons apologized for his mistake, expressing hope that the disciplinary board would consider it an honest error. This incident highlights the growing number of cases where lawyers have been disciplined for using AI tools without properly vetting their documents before submission.
Written by urgent.news from Live Mint's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
