New Mexico Supreme Court fines lawyer $5K for AI-hallucinated witnesses in murder appeal
In a murder-conviction appeal, New Mexico's highest court fined defense lawyer Stephen Aarons $5,000 and held him in contempt after his brief included fabricated witnesses and police testimony from generative AI, according to Reuters reporting carried by The Verge and The Guardian. The court cited his failure to verify factual claims before filing. Aarons said he used ChatGPT to draft a bulletproof summary, but the bench still imposed sanctions on the hallucinated material. The ruling shows courts converting LLM reliability warnings into enforceable professional penalties during active litigation. Teams shipping legal drafting assistants should treat independent fact-checking as mandatory, not optional. Reports do not identify the ChatGPT version or any verification steps attempted.
New Mexico Supreme Court fines lawyer $5K for AI-hallucinated witnesses in murder appeal
New Mexico's Supreme Court fined Stephen Aarons $5,000 and held him in contempt after his murder-conviction appeal included AI-fabricated witnesses and fake police testimony. The court acted after he failed to verify factual claims in the filing, according to Reuters reporting cited by The Verge.
Key takeaway
Unchecked generative-AI outputs in court filings can trigger contempt findings and monetary sanctions against counsel.
What happened
According to Reuters reporting cited by The Verge, New Mexico's Supreme Court fined Stephen Aarons $5,000 and held him in contempt after his murder-conviction appeal included AI-fabricated witnesses and fake police testimony.
The Guardian reported that Aarons said he used ChatGPT to create a bulletproof summary during the appeal, and the court sanctioned him for failing to verify factual claims and legal content before filing.
Evidence
The New Mexico Supreme Court fined Stephen Aarons $5,000 and held him in contempt for AI-fabricated witnesses and fake police testimony in a murder appeal.
The Verge AI · attributed
New Mexico's Supreme Court fined Stephen Aarons $5,000 and held him in contempt after his murder-conviction appeal included AI-fabricated witnesses and fake police testimony.
Stephen Aarons used ChatGPT to draft a bulletproof summary that contained fabricated police testimony and witnesses.
The Guardian AI · attributed
Stephen Aarons said he tried to use ChatGPT to create a 'bulletproof summary' during a murder conviction appeal A defense lawyer appealing his client's murder conviction submitted a legal brief containing made-up police testimony and witnesses fabricated by OpenAI's ChatGPT
The court acted because Aarons failed to verify factual claims in the filing.
The Verge AI · attributed
the court fined Stephen Aarons $5,000 and held him in contempt for failing to "verify the factual claims and legal
Why it matters
Courts are signaling that relying on ChatGPT without verification is professional misconduct in evidence-based proceedings, not an excusable tooling mistake.
Limits and uncertainties
Reporting does not specify which ChatGPT version or product tier was used, or whether Aarons relied on a dedicated legal AI tool versus a general chatbot.
Available excerpts do not detail what verification steps, if any, Aarons attempted before filing the brief.
Practical implications
Legal-tech builders should add mandatory human verification and provenance checks before any AI-generated content can enter court filings.
Law firms and operators should treat LLM drafting as a research aid that cannot replace source validation in high-stakes appeals.
What to watch
Whether other state supreme courts issue similar contempt sanctions for unverified AI-generated evidence in active cases.
Whether bar associations publish explicit guidance on ChatGPT use in appellate briefing after this New Mexico ruling.
