Field Guides / Breaking into technology / AI assurance & governance
Track G · last checked November 19, 2026
AI assurance & governance
You are skeptical, careful and a good writer, and you want to find out where AI systems fail and why.
The work: Design tests for AI systems, measure accuracy and failure modes, document risks against frameworks such as the NIST AI Risk Management Framework, and check that fixes hold.
Good fit if you’ve been: QA, inspection, compliance and intelligence analysis, and linguists who check quality.
Try it first
Experiment (2–5 hours): Test a public AI chatbot with 20 prompts and a scoring rubric, and record every failure. If this is your first one, do the foundations first.
Learn in this order
| Learn | What it unlocks |
|---|---|
| 1. AI literacy, data fundamentals, Python and SQL | Examine model behavior critically |
| 2. Test design, metrics, error categories and human review | Repeatable, evidence-based evaluations |
| 3. NIST AI RMF, privacy, security and risk documentation | Propose mitigations that matter |
| 4. Regression tests, authorized red-team exercises and reports | Technical QA and assurance evidence |
Portfolio project: An evaluation plan for an AI system, a public test set, a scoring rubric, a risk register mapped to the NIST AI RMF, and a retest after a change. You should be able to explain every part of it, reproduce the results and discuss its limits.
Credentials and degrees
Credentials: The field is new and its credentials are still settling. IAPP AIGP covers AI governance; ISACA’s CISA and ISC2’s CGRC cover audit and risk, and both expect experience. Security+ for DoD roles.
Degree: Computer Science or Data Analytics for technical evaluation; Cybersecurity for risk and compliance. Policy or linguistics electives help. How to compare schools.
Where it leads
First jobs to search for: AI evaluator, QA analyst, software tester, junior governance, risk and compliance (GRC) analyst.
Later: AI assurance engineer, AI red-teamer, AI risk or governance lead.
Reality check: Jobs titled “AI assurance” are still uncommon; much of this work sits inside QA, security and compliance teams. Search those titles too.
Related DoD work roles (orientation only; about DoD 8140): 672 AI Test & Evaluation Specialist (some DoD briefings list it as 673); 733 AI Risk & Ethics Specialist; 753 AI Adoption Specialist; 612 Security Control Assessor (adjacent).
| Occupation | U.S. 25th percentile | U.S. median | Utah 25th percentile | Utah median |
|---|---|---|---|---|
| Software QA analysts and testers | $80k | $104k | $62k | $82k |
| Information security analysts | $98k | $129k | $77k | $100k |
BLS May 2025 occupational wages via O*NET, all experience levels, rounded to $1,000. The 25th percentile is not a starting salary; first offers are often lower. No clearance, bonus or equity premium included. How to read these numbers.
Free places to start
Next moves
Enjoyed it? Stay and build the portfolio project. Enjoyed part of it more? Try a neighbor:
- Cybersecurity: Investigating threats and defending systems.
- Language & NLP: Language, translation and how AI handles meaning.