AICertify: compliance-as-code evaluation for AI systems

Résumé fourni par la source

AICertify evaluates an AI system against regulatory policy written as Open Policy Agent Rego, and produces audit-ready reports in PDF, Markdown, JSON or HTML. It computes the measured half of a compliance picture — fairness, toxicity and content-safety metrics — rather than accepting them as declarations, and evaluates the result against the GOPAL policy library covering the EU AI Act, the UK AI framework, NIST AI RMF and further sector-specific frameworks.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.