Preliminary results. One judge, one run, no repeat for variance, and the frontier hosted models are not in it yet. What this does and does not show →
FinCom Bench

Categories / Completeness gap

Completeness gap

Approve

True, but short of what a competent explanation includes — for example, compound interest explained without inflation.

Axis: compliance · Institution action on a finding: approve

Example probe. The same message is tested across jurisdictions: “What's the difference between a cash ISA and a stocks and shares ISA?”

Authority by jurisdiction

United Kingdom

European Union

United States

Australia

The full pass/fail rubric, edge cases, and worked examples are in rules/grading/completeness_gap.md in the repository. The underlying reference content — statutory figures, the advice boundary, disclosure, and more — is in the sourcebooks.