DataVisor has highlighted a significant disparity between the evolving threats posed by AI-driven fraud and financial institutions’ preparedness to counter these challenges in their latest report.
Building on this insight, DataVisor published its 2026 Fraud & AML Executive Report, which details a significant gap between the advancement of AI-based fraud tactics and the infrastructure that financial institutions have in place to combat them.
The report, compiled from data gathered through surveys of senior fraud and anti-money laundering (AML) professionals across various sectors including banks, credit unions, fintechs, and digital payment platforms, reveals that 74% of respondents view AI-driven fraud as their top concern. Despite this, 67% reported that they do not have the necessary infrastructure to deploy effective AI-based defenses.
Key Barriers to Effective Fraud Defense
The report identifies legacy systems and fragmented data as major obstacles hindering institutions’ ability to respond swiftly to emerging fraud threats. Additionally, outdated operating models are cited as significant barriers that slow down the adoption of new technologies.
As generative AI capabilities evolve, enabling more sophisticated attack methods such as deepfakes and synthetic identities, many organizations remain constrained by their current data environments and detection models, which were not designed for today’s complex threats.
However, the report also notes that financial institutions are actively addressing these challenges. Over 80% of respondents are either considering or implementing a unified approach to fraud and AML operations, with around 74% believing that achieving a single, comprehensive view of risk would greatly enhance their detection capabilities.
AI Applications in Fraud Management
The report further underscores the changing role of AI in fraud and AML operations. Traditionally associated with detection and scoring, AI is now being recognized for its potential to significantly improve investigative workflows as well. According to survey findings, 50% of executives prioritize investigator assistance through AI over more traditional use cases such as detection and scoring.
Beyond just detection, the report explores how AI can enhance operational decision-making throughout the entire fraud management lifecycle. For instance, large language model (LLM)-based AI agents are claimed to reduce false positives by 42% through optimized rulesets, while AI-assisted alert reviews and SAR narrative generation can cut review times by up to 60% and boost SAR filing efficiency by as much as 90%.
Pressure on Fraud Detection
The expansion of real-time payments, faster digital onboarding processes, and the increasing diversity of customer interaction channels are placing greater pressure on fraud and AML teams. These changes reduce the available time for detection and intervention, underscoring the need for modernized operating models and robust data infrastructures.
The 2026 Fraud & AML Executive Report offers guidance on how organizations can streamline their fraud and AML operations, enhance their data infrastructure, and operationalize AI capabilities across various stages of detection, investigation, and reporting.
