US judge fines lawyers $12,000 over AI-generated ‘hallucinated’ court filings
US judge fines lawyers $12,000 over AI-generated ‘hallucinated’ court filings
Featured

US judge fines lawyers $12,000 over AI-generated ‘hallucinated’ court filings

A federal judge in the United States has imposed financial penalties on a group of lawyers after court filings in a patent dispute were found to contain fabricated quotations and case citations generated by artificial intelligence.

Reuters reports that U.S. District Judge Julie Robinson of Kansas City ordered the lawyers representing Lexos Media IP to pay a combined $12,000 in fines, ruling that they had failed in their professional duty by submitting unverified material produced by generative AI in a lawsuit against online retailer Overstock.com.

In a strongly worded decision, Judge Robinson made clear that responsibility extended beyond the lawyer who used the technology, noting that all attorneys who signed the filings were accountable for ensuring their accuracy. “A reasonably competent attorney filing documents in court should be aware of the pronounced, well-publicized risks of using unverified generative AI for legal research and the ethical obligations associated with signing a court filing without checking it for accuracy,” she wrote.

The court found that several submissions contained “non-existent quotations and case citations” as well as misrepresented information, prompting the judge in December to order five lawyers on the case to justify why they should not face sanctions. The ruling forms part of a growing body of judicial responses to what courts describe as “hallucinated” legal authorities generated by artificial intelligence tools.

“The sheer amount of case law that has erupted over the last few years due to attorneys’ reliance on unverified generative AI research, often generating hallucinated legal authority, is staggering,” Judge Robinson said, underlining the scale of the problem confronting courts.

The heaviest penalty, a $5,000 fine, was imposed on Sandeep Seth, who admitted to using ChatGPT without verifying its output, citing personal pressures at the time. Judge Robinson further directed him to provide state disciplinary authorities with a copy of the ruling and to certify the measures his firm would take to prevent a recurrence.

In an email response following the ruling, Seth described the episode as a learning experience. “This has been an embarrassing lesson,” he said. “Firms should not use AI as a tool in any capacity without strict policies in place in order to avoid errors.”

Two other lawyers on the case, Kenneth Kula and Christopher Joe of Buether Joe & Counselors, were each fined $3,000 for failing to review the documents they signed, while local counsel David Cooper of Fisher, Patterson, Sayler & Smith was fined $1,000 for not checking the citations. None of the three responded to requests for comment.

Lawyers representing Overstock.com also declined to comment immediately on the sanctions.

The case, Lexos Media IP LLC v. Overstock.com Inc, is pending before the U.S. District Court for the District of Kansas and adds to a series of recent decisions in which courts have warned lawyers that reliance on artificial intelligence does not diminish their ethical and professional responsibilities.


Our newsletter gives you access to a curated selection of the most important stories daily. Don't miss out. Subscribe Now.

Connect With Us : 0242202447 | 0551484843 | 0266361755 | 059 199 7513 |