MEASURE-2.6 The AI system is evaluated regularly for safety risks – as identified in the MAP function. The AI system to be deployed is demonstrated to be safe, its residual negative risk does not exceed the risk tolerance, and it can fail safely, particularly if made to operate beyond its knowledge limits.
What the standard requires
AI system is evaluated regularly for safety risks – as identified in the MAP function. The AI system to be deployed is demonstrated to be safe, its residual negative risk does not exceed the risk tolerance, and can fail safely, particularly if made to operate beyond its knowledge limits. Safety metrics implicate system reliability and robustness, real-time monitoring, and response times for AI system failures.
Quoted verbatim from NIST AI Risk Management Framework (AI RMF 1.0), which is in the public domain.
Cross-framework mappings
No cross-framework mapping is seeded for this control. That is stated rather than filled in: a mapping nobody authored is worse than a gap you can see.
Integrations that carry evidence for this family
These connectors produce data relevant to NIST AI Risk Management Framework 1.0’s MEASURE family, which is the granularity CMDash publishes — there is no pre-built per-control mapping, and claiming one would be inventing it. You decide which evidence answers which control.
In CMDash this control is a row with an owner, a status, uploaded evidence kept with its full history, and an export an assessor can read.