Report: AI Adoption Fails to Shrink Fraud Teams
Despite near-universal adoption of AI in fraud and compliance departments, human teams are still growing, according to a 2026 report from SEON. A survey of over 1,000 global leaders found that rising headcounts and bigger budgets are accompanying the AI rollout. The findings suggest that AI is augmenting rather than replacing human expertise in the fight against financial crime.
- The primary driver for expanding fraud teams is the increasing sophistication and volume of financial crime, with criminals also leveraging AI. One 2024 report found that 42.5% of all detected fraud attempts in the financial and payments sector are now AI-driven. - Global money laundering losses are estimated to be at least $5.5 trillion annually, creating significant pressure on institutions to strengthen controls. In response, 89% of compliance and risk leaders report their institutions actively encourage AI use. - A key challenge hindering AI from reducing headcount is data fragmentation. A major Moody's survey found the top obstacle to AI adoption is a lack of internal expertise or skills (41%), followed by regulatory uncertainty (33%) and difficulty integrating with legacy systems (30%). - Machine learning is the most widely adopted AI technology, used by 66% of banks for fraud prevention and 65% for anti-money laundering (AML) transaction monitoring. Natural Language Processing (NLP) is the second most common, used to analyze unstructured data from sources like news articles and corporate registries. - AI tools are primarily being used to reduce the high volume of false positives generated by older rule-based systems, allowing human analysts to focus on more complex and genuinely suspicious cases. SEON clients, for example, have reported a 75% reduction in time spent on manual reviews. - Regulatory bodies are actively encouraging responsible innovation. The UK's Financial Conduct Authority (FCA), for instance, has launched a "Supercharged Sandbox" with NVIDIA and an AI Lab to facilitate live testing in controlled environments. - While the investment is significant, the potential return is high, with estimates suggesting AI-driven systems could save regulated firms up to $183 billion in annual compliance costs. A 2025 survey found that 71% of banks that invested in AI for AML have already seen cost savings, with 48% saving over $1 million in the past year alone.
