The EU AI Act read reflects the July 2026 timeline change: the main application date for Annex III high-risk systems is December 2, 2027, while transparency obligations begin August 2, 2026. Two other frameworks show up in reviews outside the EU. US customer security questionnaires cite the NIST AI Risk Management Framework, and procurement teams that ask for ISO 27001 certificates have started asking for ISO/IEC 42001. Neither is a law, and both raise the same question: what records exist, and can you produce them.

This post maps both to the runtime layer, the part that registers agents, gates their actions, and keeps the trail. Risk committees, management review, and certification audits belong to the deploying company; the runtime supplies their evidence.

NIST AI RMF: four functions, one evidence question#

NIST published the AI Risk Management Framework 1.0 in January 2023, with a Generative AI Profile following in July 2024. It is voluntary and there is no NIST certificate. Questionnaire authors use its four functions as section headings.

GOVERN asks for accountability structures: policies for AI use, defined roles, and an inventory of AI systems with assigned responsibility. The runtime artifact is the agent registry, with each agent listed under a named owner and a record of the connections it holds.
MAP asks the organization to establish the context of each system: what it does, what it can reach, who is affected. For an agent, the connection list limits what it can reach and the activity trail shows what it did with that reach.
MEASURE asks for tracking of identified risks using defined metrics. For agent fleets the measurable quantities are month-to-date model spend per agent, action volume, and the count of actions held for human decision.
MANAGE asks for the ability to respond: prioritize risks, act on them, and document the response. The runtime artifact is the gate: a gated action waits for a named person's approval or denial, and both outcomes are recorded with approver and evidence. That record is the documentation MANAGE asks for.

A questionnaire section headed "describe your AI risk management practices" gets answered function by function with those four sets of records.

ISO/IEC 42001: the certifiable one#

ISO/IEC 42001:2023 defines an AI management system (AIMS), the way ISO 27001 defines an information security management system. It is certifiable and follows the same harmonized clause structure: context (clause 4), leadership (5), planning (6), support (7), operation (8), performance evaluation (9), improvement (10), plus an Annex A control catalog.

A certification audit spends most of its time on clauses 8 and 9: show that the controls you documented operate, and that you evaluate them. Runtime records answer that evidence request:

AI system inventory and life cycle controls (Annex A's life-cycle objective). The registry shows which agents exist, who owns each, and their status.
Impact assessment inputs (the impact-assessment objective). An assessment starts from the agent's actual connections and action history instead of a design document.
Operational control (clause 8). The gate is a documented control an auditor can watch operate: a queue of held actions, each waiting on a named approver, with the decision landing on the trail.
Performance evaluation (clause 9). Spend by agent and the activity timeline are the monitoring records the clause asks the organization to review.

The certificate attests to the management system, and no runtime purchase produces one. A runtime changes the cost of the audit: reconstructing agent behavior from scattered framework logs versus pulling rows from one trail.

Where the three regimes overlap#

Side by side, the EU AI Act, the RMF, and 42001 converge on the same short list of runtime records:

Record	EU AI Act	NIST AI RMF	ISO/IEC 42001
Agent inventory with owners	Art. 9 risk process input	GOVERN	Life-cycle controls
Automatic action records	Art. 12	MEASURE	Clause 9 evaluation
Human approval with evidence	Art. 14	MANAGE	Clause 8 operation
Stop / pause capability	Art. 14 safe halt	MANAGE	Clause 8 operation
Spend and usage by system	—	MEASURE	Clause 9 evaluation

One set of records serves all three when the layer the agents run through produces them automatically, independent of each team's framework, with named humans on every decision. That is the argument for keeping enforcement and record-keeping outside the agent frameworks, made in full in the gate post.

For work routed through Deixic, the records cover agent identity, governed actions, approval decisions, results, and recorded model spend. They are inputs to the deploying company's risk process and review, and they do not establish conformity or certification. See Trust for the current security posture and SOC 2 status.