Federal Courts: Two Divergent Approaches to AI Hallucination Sanctions

Two federal judges took opposite approaches to AI-hallucinated filings the same week. Michigan federal judge Hala Jarbou (July 17) caught a DOJ brief citing Taylor v. Hott — a Sixth Circuit case that does not exist — issued a warning that hallucinated law in federal filings is unacceptable, but declined to sanction. Kentucky federal judge Thomas Cullen declined to sanction attorney Thomas Guyer over a brief containing AI-generated misquotes and incorrect citations, finding Guyer had "owned the mistake," had no history of misconduct, and was "incredibly remorseful." Separately, escalating sanction patterns tracked across other courts include a New York firm ordered to pay $10,500 and a Sixth Circuit panel fining two attorneys $15,000 each. The practical lesson for mid-sized firm leaders is not that courts are merciful — it is that first-offense contrition without a systematic verification process is not a durable defense strategy. A firm that can show a real verification step built into how AI-assisted work gets produced has a materially different story to tell than one that can only say a lawyer was sorry.