RismadarVoice Reporters
September 12, 2026
A US defence lawyer has been fined $5,000 and held in contempt after submitting a murder appeal containing fabricated witness testimony generated with the help of OpenAI’s ChatGPT.
The New Mexico Supreme Court sanctioned attorney Stephen Aarons for failing to verify information contained in the court filing, which included fictional police testimony and witnesses who did not exist.
Aarons said he used ChatGPT to summarise trial proceedings and other case materials after taking over the appeal of Oscar Renee Sandoval, who was convicted of murdering the mother of his children and sentenced to life imprisonment.
The lawyer told the court that he had assumed the artificial intelligence system would produce a reliable summary and said he did not understand the extent to which AI could “hallucinate” or generate false information.
Aarons expressed remorse, describing the incident as an honest mistake and a lesson for professionals using AI tools.
The Supreme Court said the filing contained false testimony from wholly fabricated witnesses, including a fictional account that the shooter was wearing dark trousers and a white shirt.
The justices also said Aarons showed a lack of remorse and concern for his client and ordered the matter referred to the state attorney disciplinary board for further investigation.
The case is among a growing number of incidents in which lawyers have faced sanctions for submitting AI-generated legal documents without properly checking their accuracy.
Courts in the United States have increasingly warned lawyers that AI systems can produce fabricated case citations, misquote laws and generate entirely fictional evidence.
Sandoval’s appeal remains pending and was assigned to New Mexico public defender Kim Chavez Cook on September 2.
