aichat.blog

Stanford Hai Tests Westlaw, but the Genai Results Look Worse!

OK, this story is getting into unusual territory nowOK, this story is getting into unusual territory now.
Artificial Lawyer 12:24 am on June 3, 2024

Stanford HAI researchers found that Westla, intended as the primary legal case law database, produced worse genAI-driven hallucinations than Practical Law tools in their updated study. This has raised concerns over its effectiveness for legal professionals. Updated results revealed a 33% rate of incorrect AI responses when using Westlaw compared to 17% with Practical Law.

Stanford HAI study: Identified higher error rates in Westla's genAI tools.
Westlaw vs. Practical Law: Westlaw exhibited a 33% hallucination rate, compared to Practical Laws 17%.
Updated research paper: Provides context and additional insights into the findings.
Public response expected: Awaiting TR's comments for further clarity.
AL coverage: AL has published related articles and awaits additional updates on the issue.

https://www.artificiallawyer.com/2024/05/30/stanford-hai-tests-westlaw-but-the-genai-results-look-worse/

< Previous Story - Next Story >

Stanford Hai Tests Westlaw, but the Genai Results Look Worse!

Categories