U.S. Courts Escalate Sanctions Over AI‑Generated Legal Errors

NPR reports that courts in the United States and abroad are rapidly increasing sanctions against lawyers who file briefs containing false citations and other errors generated by artificial intelligence tools, with more than 1,200 such cases tracked worldwide and about 800 in U.S. courts. Researcher Damien Charlotin of HEC Paris says penalties are rising, citing what may be a record $109,700 sanction and cost order issued by a federal court in Oregon last month against a lawyer who relied on AI‑generated material. State supreme courts are now confronting the problem directly: Nebraska’s high court in February and Georgia’s in March publicly grilled lawyers over fictitious case citations, with at least one attorney referred for discipline. Legal‑ethics experts like University of Washington associate dean Carla Wale stress that existing professional‑conduct rules already make lawyers fully responsible for verifying anything produced by AI, while some courts have begun requiring lawyers to label AI‑assisted filings—rules critics such as Above the Law’s Joe Patrice argue will become unworkable as AI becomes embedded in standard legal software. The trend underscores how generative AI is colliding with long‑standing obligations of accuracy and candor to the court, and foreshadows tougher oversight and potentially chilling effects on how U.S. lawyers adopt AI in everyday practice.