2026 Fraud Trends: Addressing Organized Crime Groups with a FRAML Approach
-
Criminals exploit data silos
-
AI and knowledge graphs are essential for future fraud detection
-
Banks must merge fraud, AML data for true investigative intelligence
The landscape of fraud detection in banking is rapidly shifting as financial institutions wrestle with increasingly sophisticated, multi-channel threats. Markus Hartmann, market intelligence expert at DataWalk, provides a forward-looking analysis of the trends and challenges banks will face as we approach 2026 via his article, “Fraud Detection in Banking: 2026 Future Trends & Predictions."
Hartmann identifies several key threats defining the 2026 anti-fraud landscape:
- Synthetic Identity Fraud: Criminals piece together real and stolen information to fabricate new identities that evade standard verification checks.
- Authorized Push Payment (APP) Fraud: AI-powered social engineering tricks genuine customers into unintentionally transferring funds into criminal accounts.
- Account Takeover (ATO) Attacks: Fraudsters leverage breached credentials to hijack customer accounts, drain funds, and perpetrate further crimes.
- AI-Powered Attacks: Generative AI automates phishing scams, deepfakes, and adaptable malware that can bypass conventional defenses.
Organized Crime Groups
Hartmann notes that financial institutions are no longer guarding against lone hackers. Instead, they’re targeting organized crime groups utilizing AI and exploiting vulnerabilities.
This is in line with one of our recent posts, where Dr. David Maimon explains the interconnected web of financial crime. While Hartmann focused mainly on digital channels, Dr. Maimon noted that any piece of information -- including a stolen check -- can be the starting point:
A stolen check isn’t just a stolen check, it's often the opening move in a chain of crimes that lead to identity theft, account takeover, or full-blown scams.
For criminals, each tactic is an interchangeable part of a wider machine.
Will 2026 be the Year of Wider FRAML Adoption?
Traditional fraud detection tools—often siloed, rule-based systems—struggle to adapt, yielding excessive false positives and failing to connect the dots across channels and accounts. This is a major reason why many financial institutions are moving towards a FRAML approach.
As noted by Hartmann:
A pivotal trend gaining momentum in larger financial institutions is FRAML - the convergence of Fraud and Anti-Money Laundering (AML) data and operations. Historically, these two functions have operated in separate silos with different datasets and objectives. However, fraud is very often the predicate offense for money laundering. By unifying fraud data (e.g., suspicious transactions, device IDs) with AML data (e.g., KYC information, SAR filings), banks can gain a holistic view of customer risk and uncover criminal pathways that would otherwise remain hidden.
With a FRAML approach, financial institutions have the ability to associate fraudulent activities such as a fraudulent check to much larger financial crime(s).
AI technologies like Anywhere On-Us Fraud or Anywhere Deposit Fraud should no longer be viewed just as tools to identify counterfeits, forgeries, and alterations, but as the "unified" approach to address financial crimes as a whole.