HPCT (SecureBio)
Science
SecureBio's Human Pathogen Capabilities Test, 100 text-only multiple-response questions on practical work with high-concern human pathogens. This measures accuracy, unlike hpct-refusal-cais, which measures refusals. Evaluated in Inspect with zero-shot prompting, usually over 10 runs. SecureBio doesn't state how refusals enter accuracy. Claude Opus 4.6 prints 80.7% with 81.9% refused, more than it could score if refusals counted as wrong, so accuracy appears to cover answered questions only and each model's denominator differs. These scores aren't comparable with other runs of the test, which is why this is a diagnostic.
Benchmark source: Securebio