AI Resilience Report forFinancial Risk Specialists

Right now, AI is showing up everywhere in financial risk work — but mostly as a helper, not a replacement. A recent industry analysis from the Global Association of Risk Professionals concluded that "It's evolution, not revolution. Instead of completely overhauling existing systems, AI is driving targeted operational enhancements, with risk professionals guiding the way to ensure AI's use is safe, ethical, and effective." The same piece reports that 53% of risk and compliance professionals are actively using or trialing AI solutions – a dramatic leap from 30% just two years ago, and that LLMs excel at tasks that previously consumed significant analyst time, including initial document review, regulatory change analysis, and drafting reports.

That maps neatly onto the most automatable tasks on the list above — data gathering, data-quality maintenance, and documentation.

A 2025 EMEA Model Risk Management survey published by Deloitte ^[1] found that two thirds of banks and insurers use models with AI or Machine Learning techniques and that 58% of banks and 30% of insurers use AI for fraud detection (AML/KYC) and 53% of banks and 37% of insurers use AI for customer experience. Deloitte also notes that 40% of participating banks are either already using or planning to use AI to enhance their overall MRM process, with frequent use cases including validation report writing, code creation, and compliance checks ^[1]. New "agentic" systems go further: BizTech Magazine reports ^[2] that tasks such as reconciliations, variance analysis, intercompany accounting and compliance checks will increasingly be handled by always-on AI agents that monitor activity, surface risks, propose resolutions and execute safely within predefined boundaries.

The hopeful news for you: GARP says practitioners anticipate that AI will change their roles rather than eliminate them entirely, with risk managers shifting toward strategic responsibilities, technical collaboration, exception handling, and AI supervision. Just as spreadsheets didn't eliminate accountants but changed the way they worked, AI will reshape rather than replace compliance functions.

Adoption is happening fast, but with real brakes. On the accelerator side, tools are now commercially everywhere, and regulators are actively trying to clear the runway. In February 2026, the U.S. Department of the Treasury released ^[3] an AI Lexicon and Financial Services AI Risk Management Framework, with a Treasury official explaining that "Clear terminology and pragmatic risk management are essential to accelerating AI adoption in financial services… These resources are designed to help institutions move faster with AI by reducing uncertainty and supporting consistent, scalable implementation." Federal Reserve Vice Chair for Supervision Michelle Bowman recently told the FSOC AI roundtable ^[4] that banks of all sizes benefit from greater efficiency, speed, and content generation, and AI will become a force multiplier for the financial system, and in the broader U.S. economy.

Costs are also dropping fast for smaller players. Deloitte reports that for small banks, AI use increased from 22% to 52% since 2023, and for small insurers, use increased from 27% to 46% since 2023, largely thanks to readily available Software as a Service (SaaS) solutions and the cost of internal development decreasing due to low-code development platforms ^[1].

But adoption is being slowed by trust, safety, and skills gaps. GARP notes that sophisticated algorithms cannot overcome poor data foundations — organizations with mature data infrastructure report far higher successful adoption rates than those with fragmented systems, and the old "garbage in, garbage out" principle still rings true. The Deloitte survey found that more than half of survey participants named transparency and explainability as a hurdle… 39% named internal skills and capabilities as a challenge to AI implementation… and 46% cited risks posed by AI ^[1].

Bowman also stressed that banks are relying on existing risk-management frameworks to guide their use of AI, and supervisors should assess whether supervisory guidance is fit for the future. And the Federal Reserve Bank of Chicago warned ^[5] about the broader picture, noting that banks need to consider the possible AI bubble tail risk—or the risk of losses due to extremely rare events from heavy AI-sector lending exposure.

The bottom line for someone considering this career: the routine, repetitive parts of risk work (pulling data, cleaning data, writing first-draft reports) are being augmented and partially automated quickly, but the higher-value work — judgment calls, client conversations, designing risk systems, and supervising the AI itself — is exactly where humans are becoming more important. GARP puts it well: complete automation of compliance processes hasn't materialized, and the near-universal belief that human oversight remains essential suggests the industry recognizes AI's limitations — rather than replacing compliance officers, AI redistributes their work toward exception handling and complex decision-making. Learning to work with these tools — not compete against them — is the move.