This is the field everyone is arguing about, and somebody has to actually govern it. The frontier labs and the companies putting AI into real workflows are hiring AI governance leads, responsible-AI managers, and model-risk specialists for work that did not exist three years ago. It runs on the new rulebooks, the NIST AI Risk Management Framework, ISO/IEC 42001, the EU AI Act, and it sits at the center of how AI gets deployed without blowing up in anyone's face. If you want to govern the systems themselves, this is where it happens.
Compliance Officer paystats as of Q2 2026
BLS put the 2024 median at $78,420. In 2026, market trackers show averages from roughly $99,000 to $120,000, with the top quartile above $160,000. Financial centers and senior governance roles sit at the top of that range.
A snapshot built from the 160 AI GRC roles currently on this page.
What these roles pay by level
Median advertised midpoint, by seniority band. Bars are to scale.
Median advertised salary midpoint by seniority level
Level
Median midpoint
Postings
Mid-level
$255k
16
Senior / lead / manager
$191k
45
Director
$230k
8
How these figures were calculated, and sources
Primary source: our own board. Every figure above is the median midpoint of the salary ranges published in the live postings on this page. Nothing is estimated, modeled, or carried over from a previous month. It is recalculated each time the board refreshes.
What is included. Only postings that publish a salary range. Hourly, weekly and monthly rates are annualised (2,080 hours, 52 weeks, 12 months). Ranges below $20,000 a year are excluded as data-entry placeholders. A seniority band is shown only when at least five postings support it; executive / c-suite is present on this page but below that threshold, so no median is published for that band.
What this is not. These are advertised ranges, not accepted offers. Advertised ranges tend to run wider than what is actually paid, and roles that do not publish a range are missing from the calculation entirely, which can bias the result upward.
For an independent benchmark, compare against the U.S. Bureau of Labor Statistics Occupational Employment and Wage Statistics for SOC 13-1041 (Compliance Officers), published annually at bls.gov/oes. BLS reports actual wages across all employers rather than advertised ranges, so its medians normally sit below job-board figures.
Where the work is
Work mode: 20 remote, 12 hybrid, 128 on-site or unstated.
Seniority mix: senior (86), mid (45), director (17), executive (12)
Employers hiring more than one: Anthropic (25), Capco (16), Openai (10), KPMG (10), Databricks (6), Mistral AI (4)
Skills these postings ask for
AI inventory and use-case intake
model risk and impact assessment
ISO/IEC 42001 and NIST AI RMF implementation
EU AI Act readiness
AI vendor and third-party assessment
How to get an AI assurance job: the full career guide for this field, covering entry routes, transferable backgrounds, certifications and salary by level.
Certifications that come up most
None of these are universally required, but they appear often enough in AI GRC postings to be worth knowing: AIGP, CIPP/US, CISA, CRISC. The certification academy covers what each one actually tests and who it is for.
AI GRC jobs: tools, skills and workflows
What tools do AI GRC jobs use?
Most of this work runs on a GRC platform holding the same objects in every organization: an inventory of AI systems and use cases, the requirements each one is subject to, the risks logged against them, the controls that answer those risks, the assessments that test the controls, and the evidence that proves any of it happened. Around that sit a model or use-case registry, a policy library, an issues and findings tracker, and whatever ticketing system the rest of the company already lives in. The tool matters less than knowing what belongs in each object and what happens when one of them changes. Our guide to GRC tools and automation skills walks through the pieces and how to show you can configure them.
What skills do AI GRC jobs require?
Process design first, configuration second. Postings on this page keep asking for the same set: mapping a process before automating it, data modeling, platform configuration, permissions, workflow logic, APIs and integrations, reporting, data quality, testing, and the patience to support the people who have to use what you built. On top of that sits framework fluency, usually the NIST AI Risk Management Framework, ISO/IEC 42001, and the EU AI Act. The cybersecurity and IT GRC career guides break the technical skills down job by job, and the career roadmaps lay out the path from analyst to Chief AI Officer.
What does the day-to-day workflow look like in an AI GRC job?
A use case arrives at intake. Someone classifies it, decides which rules apply, and either clears it or sends it for assessment. The assessment produces risks, the risks pull in controls, the controls have owners and test dates, and the tests produce evidence and findings that have to be closed on a clock. Then it repeats, because the model changes, the vendor changes, or the law does. Designing that loop, and deciding which steps stay human, is the job. Automating a confused process only makes the confusion move faster.
How do I prove these skills without holding the job yet?
Build one workflow and show it. Take use-case intake or control testing, define the records, required fields, roles, states, approvals, notifications, service levels, evidence, exceptions, reports and change controls, then explain which decisions you left with a human and why. That single artifact answers more of an interview than a certificate does. If you are studying alongside it, the 592 free certification practice questions cover CISA, CISM, CRISC and the AIGP.
Where AI GRC jobs sit next to everything else
The AI industry hires across every GRC discipline at once, so a job on this page will usually also appear on a narrower hub. AI governance jobs own the policies and oversight. AI risk jobs and model risk manager jobs handle validation and model lifecycle. AI compliance jobs and AI audit jobs cover the control and assurance lines. Data governance jobs keep the lineage every AI claim depends on. To search across all of them at once by location, function, industry and certification, use Super Search.
Two other things worth a look. If you are hiring for this work rather than looking for it, the AI governance hiring toolkit has job description templates that scope the role before you post it. And for the leadership side of the same question, how ethics and accountability actually get structured inside an organization, our sister publication covers it in the new architecture of AI governance.
Hiring for AI GRC?
We have 160 open AI GRC roles on the board right now. Reach candidates who are already searching for this role, not a general audience. Your posting appears on this page, in the job alerts, and across the GRC Careers network.
We do not track you. There are no analytics, advertising or social media cookies on this site, so there is nothing here to switch on.
Security and sign-in (always on)
Our network provider, Cloudflare, can set a short-lived security cookie if a request looks like an attack. The sign-in cookie grc_sess is set only when an employer signs in, holds a random code and nothing about you, and is deleted when you sign out. Visitors from some countries get agj_lang, which remembers the language version of the site so we do not redirect you twice.
Remember my login on this device
Off by default: you are signed out when you close your browser. Turn it on and the sign-in cookie is kept on this device for 30 days. Your choice is saved in this browser only.
Your download is on the way
Want a new one every week? Join the AI governance brief: fresh jobs, guides, and reference sheets, straight to your inbox.
Your email stays confidential. Unsubscribing is one click. At most one email a week. Your download won't be interrupted.