BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Matthew Dahl DTSTART:20250331T213000Z DTEND:20250331T224500Z DTSTAMP:20260506T225425Z UID:MathandDemoc/25 DESCRIPTION:Title: Large Legal Fictions: Detecting Legal Hallucinations in Large Langua ge Models\nby Matthew Dahl as part of NYU CDS Math and Democracy Semin ar\n\n\nAbstract\nDo large language models (LLMs) know the law? LLMs are i ncreasingly being used to augment legal practice\, but their revolutionary potential is threatened by the presence of legal "hallucinations" -- text ual output that is not consistent with the content of the law. In this tal k\, I theorize the provenance and nature of these hallucinations and discu ss methods for detecting them in LLM outputs. I then share results from th ree experiments auditing off-the-shelf LLMs and industry retrieval-augment ed generation (RAG) models\, showing that legal errors remain widespread. I conclude by emphasizing the need for empirical evidence in an age of eve r-increasing hype about AI's ability to replace lawyers and expand access to justice.\n\nMatthew Dahl is a JD/PhD student at Yale Law School and Yal e Department of Political Science. His research on AI\, judicial behavior\ , and legal citation analysis has been published in the Journal of Empiric al Legal Studies and the Journal of Legal Analysis. Before coming to Yale\ , he was a Fair Housing Fellow at Brancart & Brancart and received his BA from Pomona College.\n LOCATION:https://researchseminars.org/talk/MathandDemoc/25/ END:VEVENT END:VCALENDAR