Lakshya

Securing & Governing AI · Chapter 12

Auditing an AI system — what evidence looks like

Read this if you have been asked to audit something nobody can quite explain.

An audit needs evidence: artefacts that exist independently of anyone's assurances, that you can inspect, and that would look different if the control were not operating. AI systems produce perfectly good evidence — it is just unfamiliar, so people do not ask for it and then conclude the system is unauditable.

Control objectiveEvidence to requestWhat a failing answer sounds like
Inventoried and ownedRegister entry with named owner and tier“It's part of the platform”
Purpose and limits definedWritten intended and prohibited use“People use it for all sorts of things”
Quality is measuredEval suite, dated golden set, baseline, last N runs“It performs well”, no artefact
Changes controlledChange log tying versions to eval runs and approvalsPrompts edited in production by anyone
Data governedData-flow diagram, classification, retention, DPANo diagram exists
Access appropriateTool permission list; two-user retrieval test“It uses a service account”
Actions traceableTool-call logs: identity, arguments, result, timeOnly the final answer is logged
Humans in the loopApproval config plus a sample of real approvalsA policy saying so, no configuration
Failures detectedMonitoring thresholds, alerts, incidents closed“Users would tell us”
Explainable after the factOne reproduced decision, end to end“It's a black box”

The single best test

Pick one real output from last month and ask them to reconstruct it: exact input, retrieved context, model version, prompt version, tool calls, output, and who saw it. That one request exercises logging, versioning, retention and traceability at once. If they can do it, most of your evidence set exists. If they cannot, you have found the finding — and you did not need to understand a neural network to find it.

“It's a black box” is a finding, not an answer

The model's internals are opaque; the system is not. Inputs, retrieved documents, versions, tool calls, outputs and approvals are ordinary logs. If none exist, the issue is not that AI is unauditable — it is that this system was built without observability, and that has a straightforward remediation.

← Model risk managementThe frameworks — NIST, ISO 42001, EU AI Act →