Every compliance forum has the same anxious thread running through it right now. Is AI coming for the job. Short answer, no. Longer answer, it's going to change what the job actually is, and the regulators have already told us how.
What UK regulators are actually doing about AI
The FCA, PRA, and Bank of England aren't writing AI-specific rules. They're staying technology-neutral and applying the frameworks that already exist, Consumer Duty and SMCR, directly to AI systems. In practice that means if your firm uses AI for transaction monitoring or risk scoring, a human still has to own the outcome under SMCR. The algorithm doesn't get to be the accountable person. You do.
The FCA also launched the Mills Review in January 2026 specifically to look at how AI could reshape retail financial services, and it's running an AI Lab with a Supercharged Sandbox and an AI Live Testing pilot to help firms experiment without falling foul of rules that weren't written with AI in mind. That's not a regulator trying to shut this down. It's a regulator trying to keep up.
Why this doesn't mean fewer compliance jobs
Here's the part that gets missed in the panic. Regulators explicitly expect firms to strengthen governance and oversight of AI systems, get better at model validation and bias detection at scale, and manage the risk of relying on third-party AI and cloud providers. None of that removes a human from the loop. It adds new things a human needs to own.
That means compliance roles are shifting toward AI governance, model risk management, and regulatory liaison on AI-specific issues, not disappearing. If anything, firms need more specialist oversight capability than before, because "the AI got it wrong" is not going to be an acceptable answer to the FCA.
Who's actually at risk here
It's not compliance officers as a category. It's the generalists who never build any AI governance literacy at all. If your entire skill set is manual policy review and box-ticking with no understanding of how an AI-driven monitoring system actually makes its decisions, that's the profile getting squeezed, the same way we've talked about generalist compliance getting squeezed by specialists in other areas. AI governance is quickly becoming one of those specialisms, and it's forming right now while the market is still thin on people who genuinely understand it.
If you want to get ahead of this rather than worry about it, start building literacy in model risk and AI governance now, before every job ad requires it. Browse compliance and risk roles on Direct Compliance to see which employers are already screening for it, and position yourself before the rest of the market catches on.
Read next: SRA Warnings on AI: A staunch reminder of the need for human input, Beyond AML: are we ready for the next generation of financial crime? and Compliance Officer Salary UK (2026): Pay by Level.