Benchmarks

Citation-accuracy and reliability benchmarks for AI legal research and drafting tools, framed as risk assessment rather than product endorsement. Each entry discloses the benchmark source type (peer-reviewed academic study, vendor-published report, or independent test), the tool name and version tested, the test date, and a methodology summary, and visually separates independent findings from vendor-coordinated ones. Entries cross-link to any Risk Digest case naming the same tool. Does not contain case outcomes (Risk Digest) or how-to guidance (Workflows), though it links to both.

ToolVendorSourceHallucination rateVersion testedTest date
Analyzing Trump's 2026 Green Card Public Benefit Rule
Why the US airstrikes on Iran violated international law
Walmart Lettuce Recall Exposes AI's Role in Consumer Legal Action
Is the White House’s $200B Funding Redirect Starving Legal AI?
AI Compliance in 2026: The Enforcement Shift — How Regulators Are Moving From Guidance to Penalties
AI Compliance Software in 2026: A Buyer’s Guide to the Two-Market Split
AI crash reconstructions expose gaps in aviation safety regulation
Who is liable when an AI model escapes and hacks a third party?
Four legal controversies tied to L3Harris's Air Force One role
What Legal Claims Arise When an Airline Employee Threatens to Call ICE?
Antwerp cargo ship hydrogen fluoride leak tests three legal regimes
How Arizona's Signature-Challenge Wave Cleared Jay Feely's Path in CD1
Why Cornyn and Tillis Are Blocking Todd Blanche's Confirmation
What Rights Do Blind Travelers Have After a Flight Cancellation?
The Legal Layers Behind Sandy Brondello's 'Protected Species' Remark
Earthquake Preparedness Legal Requirements in California
How the Supreme Court's Callais Ruling Redraws the 2026 House Map
Canada-US class action risk diverges in Panasonic toaster recall
What Cruise Ship Illness Claim Attorneys Should Know About CDC Data Gaps
Chadwick Boseman brothers seek removal of estate administrator
Blogarama - Blog Directory