Showing posts with label supervising and verifying AI outputs. Show all posts
Showing posts with label supervising and verifying AI outputs. Show all posts

Tuesday, June 4, 2024

Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Tools; Stanford University, 2024

 Varun Magesh∗ Stanford University; Mirac Suzgun, Stanford University; Faiz Surani∗ Stanford University; Christopher D. Manning, Stanford University; Matthew Dahl, Yale University; Daniel E. Ho† Stanford University, Stanford University

Hallucination-Free? Assessing the Reliability of Leading AI Legal Research Tools

"Abstract

Legal practice has witnessed a sharp rise in products incorporating artificial intelligence (AI). Such tools are designed to assist with a wide range of core legal tasks, from search and summarization of caselaw to document drafting. But the large language models used in these tools are prone to “hallucinate,” or make up false information, making their use risky in high-stakes domains. Recently, certain legal research providers have touted methods such as retrieval-augmented generation (RAG) as “eliminating” (Casetext2023) or “avoid[ing]” hallucinations (Thomson Reuters2023), or guaranteeing “hallucination-free” legal citations (LexisNexis2023). Because of the closed nature of these systems, systematically assessing these claims is challenging. In this article, we design and report on the first pre- registered empirical evaluation of AI-driven legal research tools. We demonstrate that the providers’ claims are overstated. While hallucinations are reduced relative to general-purpose chatbots (GPT-4), we find that the AI research tools made by LexisNexis (Lexis+ AI) and Thomson Reuters (Westlaw AI-Assisted Research and Ask Practical Law AI) each hallucinate between 17% and 33% of the time. We also document substantial differences between systems in responsiveness and accuracy. Our article makes four key contributions. It is the first to assess and report the performance of RAG-based proprietary legal AI tools. Second, it introduces a com- prehensive, preregistered dataset for identifying and understanding vulnerabilities in these systems. Third, it proposes a clear typology for differentiating between hallucinations and accurate legal responses. Last, it provides evidence to inform the responsibilities of legal professionals in supervising and verifying AI outputs, which remains a central open question for the responsible integration of AI into law.1"