Introducing The New Frontier: Governing AI in AML Without Guidance.
Regulators have stepped back. Historical model risk management guidance was just rescinded. Prescriptive expectations are giving way to principles. The burden
now falls on financial institutions to define, adapt, and own their MRM frameworks based on their own specific set of risks.
At the same time, the updated guidance is notably silent on generative AI and agentic AI, despite many institutions already deploying or actively exploring these
technologies in AML and fraud workflows. An RFI is coming. Eventual guidance is coming. But it is not here yet. That creates both risk and opportunity.
Not every institution has the internal expertise to navigate this shift. Some, particularly smaller or resource-constrained organizations, will likely pause and wait for
clearer regulatory direction. Others with more mature capabilities may move forward, doing their best to build sound governance, controls, and oversight around
these models in real time. In the absence of guidance, something becomes better than nothing.
This is a moment. There is clear demand for practical, experience-based direction, and institutions are actively looking for it. Not theory, not frameworks built for a
different era, but honest conversation about what institutions are actually doing, what’s holding up, and where the cracks are forming. That’s exactly the
conversation we’re having at the Coalition Against Financial Crime (CAFC), and we’re not going to wait until September to bring it to you.