UK Finance Ltd.

10/14/2025 | Press release | Distributed by Public on 10/14/2025 03:24

Blog AI adoption rises in compliance, but measurement still lags

It's no secret that financial institutions are racing to adopt artificial intelligence (AI) across different areas of risk and compliance.

The opinions expressed here are those of the authors. They do not necessarily reflect the views or positions of UK Finance or its members.

According to Moody's study into AI in risk-related compliance, which surveyed 600 compliance professionals across sectors and regions, 53 per cent of respondents say they're now actively using or trialling AI, up from just 30 per cent in 2023, and awareness is nearly universal at 91 per cent. But while AI systems can now take actions like routinely screening customers, detecting anomalies, and automating Know Your Customer (KYC) workflows, returns are mixed and, in many cases, they are still hard to measure.

Of the risk and compliance professionals who participated in Moody's study, fewer than a third report a significant impact from AI to date. And among large firms (10,000+ employees), nearly a quarter say they can't assess the impact at all.

This seeming mismatch between AI adoption and measurable value might raise concern-not because the technology doesn't work, but because most firms are perhaps not measuring its impact. In fact, more than a third of compliance professionals surveyed admitted they aren't tracking performance metrics, and that figure jumps to 41 per cent among respondents from banks.

Without evidence of AI's performance, it may become harder to defend budgets, demonstrate value to boards, or establish that AI isn't an unknown quantity within critical controls.

Where AI is working - and why

Moody's study found that the organisations deriving the clearest benefits from AI are those with higher data maturity, and those that actively supervise AI's outputs rather than leaving these unchecked.

Organisations who report the highest impacts are also the ones with more structured, accessible data, and a better understanding of where AI fits within their compliance frameworks. Respondents indicated that poorly structured or siloed data led to weaker outputs, regardless of the sophistication of the AI tools in use. But when organisations invest in high-quality, well-governed data, they appear to be giving AI the foundation it needs to work more effectively.

The clearest wins today appear to be in practical, high-volume areas:

  • Fraud detection and transaction monitoring

  • Customer screening and KYC

  • Automation of repetitive compliance tasks

These are domains where AI can help detect anomalies faster than humans, process vast volumes of data in moments, and reduce the false positives that may drain compliance resources.

Fintech firms out in front, but not alone

The study also reveals sectoral divergence. Fintechsappear to be clear front-runners, with 74 per cent of professionals in that sector already using or trialling AI, compared to 50 per cent of banks and just 35 per cent of corporates.

But this gap isn't guaranteed. We're starting to see larger, more traditional institutions move from exploration to scaled adoption of AI, particularly in wealth management and professional services - 73 per cent and 60 per cent respectively. The opportunity is there, but perhaps the question remains whether firms will translate adoption into measurable value.

Compliance redefined, but not replaced

AI is already transforming how risk and compliance teams operate, but people's jobs don't seem to be going anywhere anytime soon. 96 per cent of professionals surveyed say they expect their roles to evolve as AI becomes more embedded. But only 5 per cent of professionals said they are comfortable with AI making fully autonomous decisions, most want to keep a human in the loop.

Opinion seems to be unified around the future involving human judgment and risk oversight, augmenting that with technology which can scale and spot things in data that people may miss. And compliance functions are perhaps shifting focus from doing less manual work to reallocating time to focus on complex, higher-risk scenarios where human judgment matters most.

The firms that succeed may not necessarily be the fastest adopters of AI but perhaps the ones that treat governance and explainability as an asset, with clarity, internally and externally, about what AI is doing, why it's doing it, and what value that is adding.

Area of expertise:
Payments, innovation & resilience
UK Finance Ltd. published this content on October 14, 2025, and is solely responsible for the information contained herein. Distributed via Public Technologies (PUBT), unedited and unaltered, on October 14, 2025 at 09:24 UTC. If you believe the information included in the content is inaccurate or outdated and requires editing or removal, please contact us at [email protected]