Page 31 - gs260802

Insights and Expertise

tively, AI tools need to have been fed vast repositories of or removed from processing. The answer can't depend
historical merchant data and then trained by expert ana- on a score alone. AI-driven merchant risk programs can
lysts. provide an extensive audit trail that shows how decisions
were reached.
Common problems with general-purpose AI include cold
starts, where insufficient training makes AI models un- The record makes clear which evidence shaped the out-
aware of risks outside the mainstream. Another pitfall is come and where human judgment entered the process.
gray areas, where agents can't distinguish products and Underwriting will remain an essential component of mer-
services across jurisdictions. Through adversarial testing, chant risk management, but approval is no longer the
bad actors find the thresholds of AI moderation logic, then endpoint. Payments companies need decisioning models
find "bypasses" (for example, leetspeak, which substitutes that continue to monitor for meaningful changes after on-
numbers or symbols for letters) to exploit. boarding and provide risk professionals with a defensible
basis for action.
Without sufficient context, agents can spawn false posi-
tives and negatives; without controls, they operate with- Agentic AI can strengthen merchant risk decisioning
out guardrails when it's built on merchant-specific intelligence and gov-
erned by clear boundaries. Used that way, it gives risk pro-
AI-supported risk systems need intelligence that reflects fessionals a better way to support growth while keeping
how merchant risk actually appears in the payments sys- hidden exposure from accumulating in the portfolio.
tem. This intelligence, known as context in AI systems, Dan Frechtling is a veteran risk leader serving as senior vice president
comes from regulations, card brand rules, enforcement
patterns, knowledge of bad-actor networks, and human- of Product and Strategy at LegitScript, https://legitscript.com, where
in-the-loop feedback from experienced analysts.. he drives product innovation, grows the company's expertise in mer-
chant intelligence, and strengthens partnerships with leading platforms,
Faster decisions still need a record marketplaces, and payment companies. He brings over a decade of
experience in mitigating merchant, seller, and advertising risk, having
Speed only helps when a payment company can stand held leadership roles such as president of G2 Risk Solutions and CEO of
behind its decisions. A bank, card network or regulator Boltive. Contact him via LinkedIn at linkedin.com/in/frechtling.
may later ask why a merchant was approved, restricted