Report finds financial institutions struggle to keep pace with AI-driven fraud

Financial institutions are increasingly worried about the rise of artificial intelligence-driven fraud, but many lack the technical foundation needed to defend against it, according to new research from fraud and risk management platform DataVisor.

In its 2026 Fraud & AML Executive Report, DataVisor describes a widening "AI Readiness Gap" between the growing sophistication of fraud attacks and the infrastructure financial institutions have in place to counter them. The report surveyed senior fraud and anti-money-laundering (AML) leaders at banks, credit unions, fintech companies and digital payments platforms.

Among respondents, 74 percent identified AI-driven fraud as one of their top threats, yet 67 percent said their organizations lack the infrastructure required to deploy effective AI-based defenses.

Generative AI tools are enabling new types of attacks, including deepfakes, synthetic identities, coordinated fraud rings and automated scam campaigns. At the same time, many financial institutions remain constrained by fragmented data systems, legacy detection models and governance structures that were not designed for today's real-time threat environment.

"Financial institutions are facing attackers that operate at machine speed, but many defenses still operate at legacy operational speed," said Yinglian Xie, CEO and co-founder of DataVisor. She added that closing the readiness gap will require modern infrastructure built on unified data, adaptive machine learning and operational models capable of responding to threats in real time.

Infrastructure and data barriers slow progress

The report identifies legacy infrastructure, organizational silos and outdated operating models as the main obstacles preventing institutions from responding quickly to evolving fraud tactics.

However, many organizations are beginning to rethink their approach. Eighty-one percent of surveyed institutions said they are considering or implementing a unified strategy across fraud and AML operations, while 74 percent said having a single, comprehensive view of risk would significantly improve detection effectiveness.

These findings suggest that financial institutions increasingly recognize the need to integrate data and intelligence across traditionally separate risk functions to keep pace with coordinated fraud activity.

AI's role expanding beyond detection

The research also points to a shift in how executives view the role of AI in fraud and AML operations. While AI has historically been used primarily for detection and risk scoring, many leaders now see its greatest potential in improving operational workflows.

In the survey, 50 percent of executives ranked investigator assistance as the most valuable AI use case, surpassing detection and scoring, which were selected by 44% of respondents. The shift reflects a growing focus on accelerating investigations, improving decision speed and helping risk teams act earlier on emerging threats.

DataVisor said its platform applies AI across fraud prevention workflows, including detection, investigation and reporting. According to the company, its technology can help institutions detect complex fraud patterns while reducing false positives and improving efficiency in tasks such as alert review and suspicious activity report preparation.

The report also notes that real-time payments, faster digital onboarding and increasingly diverse customer interactions are shortening the window for detecting and stopping fraud, adding urgency to modernization efforts.

The full 2026 Fraud & AML Executive Report is available at www.datavisor.com.

Notice to readers: These are archived articles. Contact information, links and other details may be out of date. We regret any inconvenience.