The Green Sheet Online Edition
October 26, 2026 • 26:10:02
When fraud looks clean, assessing risk gets harder
Some of the hardest merchant risk decisions begin with an application that looks perfectly ordinary. The business is registered, the website is functional, ownership information appears consistent and transaction activity falls within expected ranges. Nothing gives underwriters a clear reason to reject the application, or risk teams an obvious reason to stop processing. The data signals don't trigger concern when viewed within a limited scope.
Cases like these are difficult because the individual signals may not look risky at all. They also raise an important question as payments organizations rely more heavily on AI for fraud detection and merchant monitoring. What happens when bad actors have manipulated the data a system is evaluating to look normal?
Fraudsters know what risk systems look for
A recent federal case illustrates the problem. In August 2026, the owner of payment processing company CB Surety pleaded guilty to conspiracy to commit bank fraud for his role in a scheme that helped high-risk and fraudulent merchants obtain and maintain merchant accounts (see tinyurl.com/4hu48tje).
According to the Justice Department, participants created sham companies, recruited straw owners, and developed fake websites and contact information to make merchants appear legitimate. The operation went further by generating large numbers of small transactions that artificially lowered merchants' chargeback rates. More than $111 million in transactions were processed through the scheme.
While this is an extreme example, the case is a reminder that the signals payments organizations use to evaluate merchants can themselves become targets for manipulation. Chargeback rates, business information and transaction patterns remain valuable indicators, but by themselves, they are only data points that lack greater context.
A low chargeback rate typically provides a risk team with useful information about a merchant. It means something different when artificial transactions have been added specifically to change that ratio. The same is true of a complete merchant application. Registration documents and a professional-looking website may satisfy individual checks while still presenting an inaccurate picture of the business as a whole.
That is where sophisticated risk becomes difficult. Each piece of information may appear reasonable on its own, while the connections between those pieces suggest something deserves more attention.
AI can surface the problem without settling it
AI has an important role in finding those connections because risk teams are evaluating signals across merchant applications, transaction activity, websites and other sources that are difficult to piece together manually. Automated systems can identify inconsistencies and bring related evidence together for review. The harder part comes when there is no familiar fraud pattern to identify.
Sophisticated cases often involve merchants that have already passed several standard checks. Their documentation may be legitimate, their activity may fall within normal ranges, and their websites may contain nothing that clearly violates a policy. The concern may become visible only when several smaller discrepancies are considered together.
A business may describe itself one way during underwriting but gradually move into a different product category. Two companies presented as unrelated may have connections that are difficult to explain. Information found elsewhere online may conflict with what the merchant told its payment provider.
None of those signals on their own necessarily means a merchant is fraudulent. Legitimate businesses change, and inconsistencies can have reasonable explanations. That uncertainty is exactly why human judgment remains a critical component in the risk-assessment loop.
AI can help an analyst find the information that deserves attention, but the analyst still has to decide what that information means. That requires understanding the merchant's business model, assessing whether the explanation is credible, and determining whether the evidence is sufficient to warrant additional scrutiny or action.
A merchant account is only part of the picture
Merchant risk rarely exists within the boundaries of a single account. A business does not exist only within a payment processor's records. It may operate several websites, advertise across different platforms, or maintain other online properties that provide additional context about what it sells and how it conducts business. Looking at signals across the commercial internet can reveal relationships that are difficult to see through transaction data alone.
For example, a merchant offering innocuous products on their website may use social media forums to link to hidden subpages that market violative products. Or, a seemingly compliant merchant might offer sales of violative products off platform, through messaging apps or by directing customers to other websites. This broader context is difficult to capture when automated systems are trained to review only certain data points.
Where automation ends and judgment begins
A comprehensive risk program gives teams enough information to make a call they can explain and defend. That becomes harder when the evidence is mixed or when a merchant falls somewhere between clearly acceptable and clearly problematic.
In those cases, a simple pass-or-fail approach is not enough. Risk teams need to look at how the available information fits together, what has changed and whether the concerns are significant enough to justify further action.
As fraudsters get better at making individual signals look legitimate, effective risk management will depend less on any single indicator and more on the ability to connect information across the merchant's broader footprint. AI can make those connections easier to find, but expert investigators still provide the context and judgment to determine what they mean. 
Dan Frechtling is a veteran risk leader serving as senior vice president of Product and Strategy at LegitScript, https://legitscript.com, where he drives product innovation, grows the company's expertise in merchant intelligence, and strengthens partnerships with leading platforms, marketplaces, and payment companies. He brings over a decade of experience in mitigating merchant, seller, and advertising risk, having held leadership roles such as president of G2 Risk Solutions and CEO of Boltive. Contact him via LinkedIn at linkedin.com/in/frechtling.
Notice to readers: These are archived articles. Contact information, links and other details may be out of date. We regret any inconvenience.
