Article

A map of AI agent governance techniques

Thirteen families of controls for AI agents, where each one sits in the request path, and which risk it actually reduces.

Published 4 min read
Contents

An AI agent is a model that can take actions: call an API, run code, send an email, move money. Governance is the set of controls that decide which of those actions happen, under whose authority, and with what record. This article maps the main techniques so that later pieces can go deep on each one.

The short version: most real risk comes from the combination of untrusted input, access to sensitive data and the ability to act. Each technique below breaks one link in that chain, and none of them is enough alone.

Where the controls sit

It helps to picture a single request travelling through an agent. Untrusted content comes in, the model plans, a tool call goes out, and a result comes back. Controls attach at each step, and a few cut across all of them.

Figure 1. Where each family of controls attaches to an agent's request path. The policy gate is the one place every action has to pass through.

Controls on the action path

Runtime policy enforcement

A policy gate sits between the agent and its tools and evaluates every proposed call against explicit rules before it runs. It is deterministic, testable and independent of the model’s reasoning, which is why it is the backbone of most serious deployments. The companion piece, Runtime policy enforcement for AI agents, covers implementations in detail.

Sandboxing

When an agent runs code or touches a file system, sandboxing limits the blast radius: an isolated container or micro-VM, a read-only mount, and network egress restricted to an allowlist.

Human in the loop

Some actions should wait for a person. The design question is not whether to ask but what to show: a reviewer who sees “approve tool call?” will click yes; a reviewer who sees the amount, recipient and reason can actually decide.

Controls on inputs and outputs

Prompt injection defence

Any text the agent reads can contain instructions. Filters and classifiers help at the margin, but the durable defences are architectural: keep untrusted content away from privileged tools, and treat model output that was influenced by untrusted content as untrusted too.

Output guardrails

Validate what the agent produces before it reaches a user or another system: schemas for structured output, allowlists for URLs, checks for personal data.

Memory integrity

Long-term memory turns a one-off injection into a persistent one. Record where each memory came from, let users see and delete it, and do not let content from untrusted sources write to memory that privileged steps later read.

Cross-cutting controls

ControlWhat it answersTypical mechanismMain risk reduced
Agent identity and accessWho is acting, on whose behalf?Per-agent service identities, OAuth on-behalf-of tokens, short-lived scoped credentialsOver-privileged agents, unattributable actions
Containment and circuit breakersWhat happens when it goes wrong at speed?Spend and call budgets, rate limits, anomaly-triggered shutdownRunaway loops, cost blow-ups, cascading failures
Observability and auditWhat did it do and why?Structured traces of prompts, tool calls, policy decisions and approvalsUndetected misuse, failed audits
Multi-agent trustCan this agent trust that one?Signed delegation, capability narrowing at each hopConfused deputy across agents
Supply chainIs this tool what it claims to be?Pinned versions, provenance checks, review of tool descriptionsMalicious or changed tools and MCP servers
Evals and red teamingDoes it behave as intended?Scenario test suites, adversarial testing, regression on each changeSilent behaviour drift

How the frameworks map onto this

Regulation and standards rarely name techniques; they ask for outcomes. Roughly:

  • EU AI Act. For high-risk systems, the Act requires record-keeping (automatic logging), human oversight and technical robustness. Observability, human-in-the-loop and policy controls are how teams evidence those requirements.
  • NIST AI RMF. The Govern, Map, Measure, Manage functions are a lifecycle, not a control list. The techniques above live mostly in Manage, with evals in Measure.
  • ISO/IEC 42001. A management-system standard: it asks you to identify AI risks, choose controls and show they operate. This map is a candidate control set for that exercise.

Where to start

If you deploy one agent next quarter, the order that buys the most risk reduction for the least effort is usually:

  1. Give the agent its own identity with the narrowest credentials that work.
  2. Put a policy gate in front of every tool call, starting with a default-deny allowlist.
  3. Log every tool call and policy decision in a structured, queryable form.
  4. Add budgets and a kill switch.
  5. Add human approval for the small set of actions that are irreversible or expensive.

Everything else, from injection classifiers to multi-agent delegation, builds on these five.

Get the next piece by email

New articles and whitepapers on controlling AI agents, with a short note on what changed in regulation and tooling. No promotions.

One email every two weeks. Unsubscribe in one click.