Visa/Featurespace: Why Banks Need a Connected Fraud Ecosystem That Includes Checks
- Visa is fighting fraud at the network level—banks must match that scale
- Criminals share data across rails; fraud defenses need to do the same
- OrboGraph's check image intelligence closes a critical blind spot for banks
In a recent interview, James Mirfin, SVP, Head of Risk and Security Intelligence Solutions at Visa, spoke with PYMNTS CEO Karen Webster about Visa’s network-level fraud strategy, which points to a future where banks, networks, fintechs and check processors must operate as a connected fraud ecosystem, not a collection of silos.
Fraud is an AI Powered Business
Visa’s James Mirfin describes fraud as a “business, an economy” that has professionalized around AI, automation, deepfakes and voice cloning. Criminals run scalable operations that test and resell credentials, move between channels and monetize stolen data wherever controls are weakest—including through check fraud. They also operate their own data-sharing networks, where stolen identities, cards, accounts and checks circulate freely.
This means financial institutions are no longer fighting isolated card, ACH, or check events; they are competing with an always-on, data-rich criminal ecosystem. The only effective response is an even more coordinated data-sharing environment that spans every rail and every payment instrument, with checks fully in scope.
From Silos to an Ecosystem Model
Mirfin notes that many banks still make decisions in silos, with separate tools and teams for provisioning, authentication and authorization. Each unit only sees part of the picture, which leads to false positives, false declines and missed cross-channel patterns.
Visa’s approach turns that on its head by embedding intelligence across the full payment lifecycle and at the network level. Data from provisioning, authentication and authorization is combined, then pushed back to issuers. For example, Visa's recent acquisition of Featurespace enables them to not only leverage their AI-powered behavioral analytics to assess the activities and behavior of accounts, but also leverage OrboGraph's Image Forensic AI to scrutinize the images of checks.
By connecting check fraud data with card and digital payment fraud platforms, institutions can:
- Correlate suspicious check transactions with account takeover and card-not-present activity.
- Spot multi-channel mule behavior across check deposits, P2P inflows and outgoing transfers.
- Update defenses rapidly as new check fraud patterns emerge, without massive IT projects.
Mr. Mirfin’s goal is to make “the cost of trying to enact fraud higher than the payback.” By combining Visa’s network intelligence with powerfully image forensics for check fraud detection in a shared, data-rich ecosystem, financial institutions can raise that cost dramatically—across every rail, every channel and every single check.