Download Framework
MLCommons / AILuminate icon
https://mlcommons.org/
mlcommons-ailuminate

MLCommons / AILuminate

Industry-standard safety benchmark testing AI chat systems with 24,000 prompts across 12 hazard categories. Five-point grading scale.
Type:
Industry
Domain:
Academic/Professional
Coverage:
Performance & ReliabilityBias & FairnessSafety & Reputational Harm
Content:
0 Risks0 Controls
Version: 2025

Framework Definition

Risks and controls associated with the framework

Assessment Layer

Concrete evaluations linked to controls to assess pass or fail
No evaluation mapping defined yet.