Some law firms reap fees for opponents’ AI errors, but one lawyer who failed to detect them got zilch

House
Each day Information
Some legislation companies reap charges for opponents’ AI…

Synthetic Intelligence & Robotics

Some legislation companies reap charges for opponents’ AI errors, however one lawyer who didn’t detect them bought zilch

By Debra Cassens Weiss

September 25, 2025, 2:40 pm CDT

Judges are imposing hefty fines on attorneys who use defective case citations in authorized briefs which are doubtless generated by synthetic intelligence. The sanctions often are paid to legislation companies that expended time because of the errors—however not in a single latest case in California.

In Puerto Rico, a number of companies—together with Paul, Weiss, Rifkind, Wharton & Garrison and Sidley Austin—will share $24,492 in authorized charges after the opposing counsel filed motions with no less than 55 faulty citations, report Reuters, Law360 and the Volokh Conspiracy.

The solo practitioners who filed the motions had denied that the errors had been prompted by way of AI, “however the sheer variety of inaccurate or nonexistent citations suggests in any other case,” stated Chief U.S. District Choose Raúl Arias-Marxuach of the District of Puerto Rico in a Sept. 23 opinion.

The companies engaged on the case had initially sought almost $88,000 in charges, however Arias-Marxuach trimmed the quantities, Law360 factors out.

The 2 attorneys who had been fined are Ibrahim Reyes of Reyes Attorneys and Jose R. Olmo-Rodriguez of the Olmo & Rodriguez Matias Legislation Workplace. Neither instantly responded to the ABA Journal’s request for remark.

In California, a lawyer should pay $10,000 for together with pretend quotes and off-topic instances generated by AI in appellate briefs, report CalMatters and LawSites.

The sanction shall be paid to the court docket, the California Courts of Attraction’s Second Appellate District stated in a Sept. 12 opinion. The appeals court docket declined to award legal professional charges to the opposing counsel as a result of they didn’t detect the errors.

The case “provides an intriguing dimension to the rising physique of AI-hallucination-sanctions instances, elevating the query of a lawyer’s obligation to detect fabricated, AI-generated citations—not within the lawyer’s personal filings however in an opponent’s,” LawSites experiences.

The sanctioned lawyer within the California case is Amir Mostafavi. He instructed CalMatters that it’s unrealistic to anticipate attorneys to cease utilizing AI. However till AI stops producing bogus info, attorneys should proceed with warning.

“Within the meantime, we’re going to have some victims, we’re going to have some damages, we’re going to have some wreckages,” Mostafavi stated. “I hope this instance will assist others not fall into the opening. I’m paying the value.”

Write a letter to the editor, share a narrative tip or replace, or report an error.