The new compliance partnership: AI, tech and the evolution of humans in the loop

Robin Cotterill, Chief Compliance Officer, Carne Group

Financial crime compliance has become a compelling use case for technology. Expanding sanctions regimes, increasingly complex investment structures and regulatory divergence are creating more data, more decisions and more work for compliance teams. Combine all of that with increasing volumes of private assets transactions requiring KYA, and it’s easy to see why manual processes alone can no longer keep pace.

Technology is therefore becoming essential to keeping up with regulatory requirements. But the opportunity is not simply to automate more of compliance. It is to change how technology and people work together.

AI is accelerating that shift, taking on more of the data-intensive work within compliance. The question is where the handoff between humans and systems happens – and how firms ensure the human in the loop remains focused on the decisions that technology cannot make.

Financial crime compliance has become too complex for traditional operating models

The scale and pace of change in financial crime compliance are making manual processes increasingly difficult to sustain. Sanctions are a good example. What was once a relatively stable list-screening obligation has become a high-volume, high-change compliance discipline. LSEG’s Global Sanctions Index has risen by 446% since January 2017, with its latest data showing just under 80,000 sanctioned persons globally.¹

But volume is only part of the challenge. Sanctions have also become more dynamic and fragmented, with different jurisdictions taking different approaches to target selection, timing, legal thresholds and ownership-and-control tests. Geopolitical developments add another layer of change, triggering rapid shifts in requirements.

More broadly, compliance teams are dealing with greater volumes of information and more frequent regulatory developments across jurisdictions. Against this backdrop, technology is becoming more than a means of improving efficiency. It’s now near-fundamental to keeping pace in an increasingly complex and nuanced compliance environment.

AI is changing the role of compliance professionals

Technology can reduce the burden of manual processes across the compliance function, from screening and workflow management to data handling and regulatory monitoring. AI takes this further where large volumes of information need to be captured, interrogated and assessed.

Its greatest value lies in its ability to capture, validate and rationalise large quantities of data quickly. That is revolutionising how compliance teams spend their time, speeding up information gathering and allowing specialists to focus on the areas where their expertise matters most.

The same principle applies to regulatory change. AI can support continuous regulatory horizon scanning and mapping, helping firms keep track of developments across jurisdictions and respond more effectively as requirements evolve.

That’s crucial because the regulatory environment is far from static. Even efforts to create greater harmonisation, such as the EU’s new Anti-Money Laundering Authority (AMLA) and common rulebook, sit alongside local implementation and continued divergence between jurisdictions. AI becomes increasingly useful, therefore, in finding, organising and processing that information in real-time. In turn, the role of compliance professionals is likely to become more focused on the areas where human expertise adds most value: interpreting information, challenging assumptions and making informed decisions, supported by AI.

Now, the challenge is navigating the handoff between technology and humans

The next question is where technology’s role should end and human expertise begin. Technology can speed data collection, assess initial risks and deliver useful, concise data sets to subject matter experts faster than ever before. It can help firms process information at a scale that would be difficult to achieve through manual approaches alone.

But the output is not necessarily the final answer. Complex onboardings and transactions still require experienced professionals to understand what the data is telling them, assess the bigger picture and make the right judgement calls.

That distinction is particularly important where decisions depend on context rather than simply identifying a potential match or flag. Technology can surface the relevant information, but people need to interpret it within the context of the transaction, the jurisdictions involved and the wider risk picture.

This is just as important for people as it is for technology

The pressure on compliance teams is already increasing. AML teams are being asked to do ever more complex work, often without a corresponding increase in resources. There are commercial pressures, too. Deal teams do not want due diligence and KYA checks to hold up execution, while MLROs need to ensure that financial crime risk is properly managed. That can create friction between the need for thoroughness and the need to move at pace.

It’s also harder than ever to find experienced compliance professionals. Increasing competition for talent is leaving a gap for AI to fill in making scarce expertise go that bit further. By taking on more of the lower-value, manual processing, automation can allow experienced specialists to concentrate on the complex cases and decisions where their expertise matters most. That creates the potential to absorb greater volumes of activity without simply relying on more manual processes.

Where humans add the most value

There is no question that AI and automation will take on more of the operational workload across AML, sanctions and KYA. The opportunity is not to see technology as a substitute for human expertise, but to combine the two in a way that allows managers to dynamically respond to growing transaction volumes and regulatory complexity.

That requires the technology to be calibrated effectively, so it captures what matters without drowning the business in white noise. It also requires investment in the people behind it: well-trained and well-resourced SMEs who can challenge what the technology produces and make the decisions that ultimately determine how financial crime risk is managed.

The human in the loop isn’t disappearing. Their role is evolving – from processing information to interpreting it, challenging it and making the decisions that technology still can’t.

For more on technology’s growing role in compliance, register your interest for our next RegsRadar webinar.