Judge Dismisses Case Over Lawyer's Fake AI Citations

A federal judge in New York has dismissed an entire case after attorney Steven Feldman repeatedly filed court documents containing fabricated legal citations created by AI tools. The extraordinary sanction marks one of the strictest judicial responses yet to AI misuse in courtrooms.

District Judge Katherine Polk Failla grew suspicious when one of Feldman's filings suddenly shifted from containing grammatical errors to featuring flowery prose. The document included an extended quote from Ray Bradbury's "Fahrenheit 451" and compared legal work to gardening and leaving "indelible marks upon the clay."

Another passage referenced ancient libraries and a Bible verse about divine judgment, prompting the judge to raise her eyebrows. The writing style was so dramatically different from Feldman's other filings that Judge Failla concluded AI had written it, despite the lawyer's denials.

Feldman insisted he wrote every word himself, claiming he'd read the Bradbury book years ago and wanted to add personal touches. He admitted using three different AI programs—Paxton AI, vLex's Vincent AI, and Google's NotebookLM—but only to check citations, not to write the documents.

The judge wasn't convinced. She wrote that it was "extremely difficult to believe" AI hadn't drafted those sections, accusing Feldman of dodging the truth and trying to obscure his AI misuse.

Feldman defended his reliance on AI by explaining he couldn't afford expensive legal database subscriptions and struggled to access the library due to short hours and family commitments. With multiple cases and his children's graduations to attend, he said AI seemed like a practical solution for verifying citations.

Why This Inspires

While this story shows what not to do, it's actually good news for justice. Judges are paying attention and holding lawyers accountable for accuracy, protecting the integrity of our legal system.

Hundreds of lawyers have faced consequences for submitting AI-generated fake citations, with fines ranging from $150 to $85,000. Some law firms now threaten termination for citing fabricated cases, and lawyers are voluntarily taking leaves of absence to rebuild their credibility.

The message is clear: technology can be a helpful tool, but there's no substitute for human responsibility and thorough verification. Courts are adapting to the AI age by setting firm boundaries that protect everyone who relies on the legal system for fair treatment.

This ruling reinforces that shortcuts undermine justice, and that's something worth celebrating.