Observed arrival · 2026-09-04
The Citation Record: a legal-AI benchmark with a paper trail
An independent quarterly benchmark testing whether legal AI systems cite the law accurately, not merely whether the cited cases exist.
○Open
⊠Login
$Paid
†Ads
✦Pretty
●Pro
◎Niche
◉Human
⚑Risk
ƒJS
Why it surfaced
The project separates six citation failures, including fabricated cases, wrong holdings, overruled authority, and nonbinding jurisdictional sources. Its unusually candid disclosures include testing the publisher's own tool under the same rubric and publishing vendor responses unedited.
An independent quarterly benchmark testing whether legal AI systems cite the law accurately, not merely whether the cited cases exist.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use
One card from the complete issue