Skip to content
Submit incident
AII-1001-95343
Under investigationA regulator, court or police has opened a formal inquiry and reached no finding yet.Supporting sources are pending verification.

Judge considers sanctions over AI-generated false citations in Alabama prison case

Attorneys at a law firm hired to defend Alabama's prison system submitted two court filings containing five false case citations generated by ChatGPT without verification. A federal judge is considering sanctions and has scheduled hearings on whether the lawyers should face penalties.

21 May 2025 Β· Butler Snow Β· Reliability and hallucination
21 May 2025
AII-1001-95343
Quick facts
Category
Reliability and hallucination
Country
United States
Date
21 May 2025
Drafted by
Raidu Incident Detection

What happened

Attorneys at Butler Snow LLP, a firm hired to defend Alabama's prison system, used ChatGPT to research case law for two court filings in federal court. Partner Matt Reeves used the AI system to generate supporting citations but did not verify the information before adding it to the filings, which were signed by four attorneys including Reeves.

U.S. District Judge Anna Manasco identified five false citations across the two filings. The citations were described as "hallucinations" β€” incorrect citations generated by the AI system that did not correspond to actual cases.

Reported impact

Judge Manasco held a hearing on May 21, 2025 to question the attorneys about the filings and is considering sanctions including fines. Butler Snow apologized in a response to the judge, stating that using ChatGPT to obtain legal authority without verifying sources was 'against good judgment and firm policy' and had 'no excuse'.

Index analysis

Judge Manasco noted that there have been broad national warnings from courts about the use of artificial intelligence to generate legal filings because of the potential for inaccuracies. The judge conducted independent searches for each allegedly fabricated citation and found none to exist.

Governance controls

Butler Snow stated that relying on ChatGPT to research case law without verification violated firm policy. Matt Reeves stated he alone was responsible for the false citations and asked the judge not to punish his colleagues.

Classification

Category
Reliability and hallucination
Industry
Government, security and defence
Country
United States
Harm status
Potential harm
Intent
Unintentional
Timing
Post-deployment
Harm types
Reputational
Harmed
Business
Failure modes
Fabricated citation, Hallucination, Overreliance
Consequences
Litigation
AI systems
ChatGPT

Sources and evidence

Sources not yet published

This record was researched and written by the Index from original reporting. Its sources are still being verified by an editor, and they are listed here once that check is complete. Until then the record carries no reviewer.

How sources are verifiedSuggest a source for this incident