AI System Performance: Measuring performance on your own population

NIST AI RMF MEASURE 2.3 sets the expectation for AI system performance. The text is brief; the evidence it implies is not, and the difference between the two is where most remediation time goes.

What the framework says

AI system performance or assurance criteria are measured qualitatively or quantitatively and demonstrated for conditions similar to deployment setting(s). Measures are documented.

That is the text of NIST AI RMF MEASURE 2.3. Read it closely: it describes an outcome to achieve, not a product to buy, which is why two practices of the same size can satisfy it in different ways and both be right.

What this looks like in practice

In an assessment, NIST AI RMF MEASURE 2.3 is not one question. It resolves into several, and each one is really asking for different evidence. These are among the questions asked about it:

• Is this AI system's performance measured and demonstrated under conditions similar to its deployment setting?
• Have the capability and accuracy claims made for this generative AI system, by its vendor, the project team or user-facing material, been checked with systematic, empirically validated methods rather than narrow or anecdotal spot checks?
• Before release, was this AI system tested against defined acceptance criteria, using data and conditions representative of where it will operate, with results and sign-off recorded?

Ownership usually sits with the model owner, though the evidence is often produced by someone else, which is where the trail breaks. A control that works but has no owner tends to stop working the month the person who quietly maintained it changes roles.

Who this applies to

NIST AI RMF is voluntary rather than binding, and that changes how it should be used. Adopting it does not create a legal obligation, and declining it is not a violation. It earns its place when a customer contract names it, when it gives structure to AI governance work you already owe under a law or a contract, or when a recognized practice is worth having on record. Treat it as a way of organizing that work, not as a second regime.

Building it, in order

The sequence below is the order that produces evidence as a by-product rather than as a separate documentation exercise:

1. Define performance or assurance criteria.
2. Test under conditions representative of deployment.
3. Record results against the criteria.
4. List the capability and accuracy claims made about the system, whether by the vendor, the project team or user-facing material.
5. Evaluate each claim with a documented method, representative test sets, benchmarks relevant to the use case, expert grading, using realistic prompts and data.
6. Record factual error and confabulation rates together with the limits of the evaluation.

Where else this maps

The questions behind NIST AI RMF MEASURE 2.3 also map to NIST AI 600-1 2.2 and ISO/IEC 42001 A.6.2.4, which are voluntary guidance and standards rather than law; following them does not by itself establish compliance with any law. That matters for scoping: the same evidence can be written once and referenced from each, rather than rewritten per framework.

Evidence an assessor asks for

For this outcome the evidence type that satisfies it is a report. The distinction matters more than it looks: a policy states what you intend to do, a procedure states how, and a record proves it happened on a date. Auditors ask for all three, and a practice that has written the first two often has nothing for the third.

Date every document and keep the superseded versions. The framework sets no retention period, so this is our recommendation rather than a requirement: keep each version for as long as the system it describes is in use and for a period after it is retired, because a question about a past decision is answered from the version that was in force when it was made.

Where this usually goes wrong

The usual gap is a control that exists in practice and nowhere in writing. Someone does the work, reliably, and it has never been written down, so the organization cannot demonstrate it and cannot notice when it stops. The fix is not more control, it is a dated record and a named owner.

What to do about it

• Produce the evidence, not the intention. For NIST AI RMF MEASURE 2.3 that means a dated report with the period it covers and who reviewed it. If the only evidence is that someone knows how this works, you have a practice and not a record.
• Put a name on it. The model owner should be recorded as the owner. Where the work is done by someone else, say so in the document: the split between who owns it and who performs it is where the trail usually breaks. A control with no owner stops working the month the person who quietly maintained it changes roles, and nobody notices.
• Set a review cadence and keep the evidence of each review. An annual date with a dated output is a common floor; what fails an assessment is a control implemented once and never re-checked against a changed environment.
• Do not treat this as a legal obligation. It is a voluntary framework. Adopt it where a customer contract or a regulator's guidance names it, or where it gives structure to AI governance work you owe for other reasons, and say in your own documentation which of those reasons applies.

None of this makes AI system performance harder than it is. The outcome asks for a decision, an implementation, and a record. Organizations that struggle at assessment time are rarely the ones without controls; they are the ones that cannot show when a control started or who owns it now.

Related outcomes

• AI Metrics Selection: Choosing metrics that reveal harm rather than only accuracy
• AI Security and Resilience: Testing an AI system against adversarial input
• AI Impact Assessment: Assessing impact on patients before deployment

References

• MEASURE 2.3, in NIST AI Risk Management Framework 1.0 (NIST AI 100-1). This is a voluntary framework, not a legal requirement. The quoted statement is NIST's own, from the Core tables of NIST AI 100-1.
• Crosswalk references in this post are the mappings in our question bank, and are not verified against the published text of those standards.
• This post describes what a framework outcome asks for and what evidence documents it. It is not legal advice, and your counsel owns the conclusion. State law may impose additional obligations of its own.