US District Judge Brantley Starr of the Northern District of Texas issued the requirement on Tuesday, warning lawyers that AI tools can create fake cases and that he may sanction them if they rely on AI-generated information without first verifying it.

Starr said in an interview that he devised the requirement to put lawyers on notice that they can't just trust AI-generated information without verifying it.

He said he considered banning the use of AI in his courtroom altogether but decided against it after conversations with UCLA School of Law professor Eugene Volokh.

The judge also noted that he and his staff will avoid using AI in their work altogether.

The order said all attorneys must attest that either no portion of the filing was drafted by generative artificial intelligence or that any language drafted by generative artificial intelligence was checked for accuracy.

The statement noted that while AI platforms are powerful, they are prone to hallucinations and reliability and bias are other issues to consider.