Ballerine's Fraud Detection API Targets 60-Second Merchant Risk Scoring

Ballerine has launched an agentic detection system that flags high-risk merchants and coordinated fraud networks in under 60 seconds, a response to what payments industry figures are calling an "AI scamdemic" of automated merchant fraud. The San Francisco company's new Scam & Fraud Detection API went live on 21 April 2026, targeting a structural weakness in how acquirers and payment providers currently identify bad actors before money moves.

The tool departs from conventional fraud defence in one important respect: rather than scoring individual transactions or onboarding applications in isolation, it analyses the broader digital footprint around a merchant, including behavioural signals, ownership structures, and ecosystem connections between related entities. That shift reflects a broader recognition across the payments sector that fraud rings now operate as coordinated networks spanning multiple identities, geographies, and acquirers.

Why is merchant fraud accelerating now?

Generative AI has collapsed the cost and timeline of building convincing fake businesses. Scam operations that previously required months of setup, including registered entities, functional storefronts, and transaction history, can now be assembled in days. The result is a rising volume of synthetic merchants that pass traditional onboarding checks while concealing coordinated activity underneath.

The payments industry has been tracking this shift for several years. Visa's 2024 Biannual Threats Report flagged a sharp rise in enumeration attacks and automated fraud schemes, while Mastercard acquired AI fraud-detection firm Brighterion and later Recorded Future in a $2.65 billion deal to bolster its threat intelligence capabilities. The common thread across these moves is a recognition that rules-based systems built for a slower era of fraud are being routinely bypassed.

Cihat Fitzgerald, Ballerine's Chief Risk Officer and former Head of Ecosystem Risk at Visa, frames the problem as one of structure rather than volume. "What we are seeing is not just more fraud, but more structured fraud," he said. "These are coordinated systems designed to move across acquirers and geographies. That creates pressure across the entire payments ecosystem."

What does the Ballerine API actually detect?

According to the company, the Scam & Fraud Detection API surfaces four categories of risk that typically evade single-merchant review: clusters of related entities operating under different names, discrepancies between declared business activity and actual online behaviour, early signals of transaction laundering, and high-risk activity embedded inside otherwise legitimate storefronts.

The underlying approach is closer to network analysis than traditional transaction monitoring. By mapping connections between merchants, acquirers can identify synthetic identities and deceptive storefront networks before they generate chargebacks or regulatory scrutiny.

Noam Izhaki, co-founder and CEO of Ballerine, pointed to the timing gap as the core commercial problem. "If it takes days to determine whether a merchant is risky, that gap is already being exploited at scale," he said. "Our Scam & Fraud Detection API solution can determine a malicious merchant in less than 60 seconds and block fraudulent transactions before they happen."

How does this fit into agentic commerce?

The launch arrives as the payments industry prepares for agent-driven transactions, where software agents autonomously discover, evaluate, and purchase from merchants on behalf of end users. OpenAI, Google, and Stripe have all signalled ambitions in this direction, with Stripe launching its Agentic Commerce Protocol in 2025. In that model, the ability to verify merchant legitimacy in real time becomes a prerequisite for trust rather than a back-office compliance task.

The launch is backed by research from The Synthetic Merchant Economy: How Fraudulent Merchants Are Infiltrating the Payment Ecosystem, a book authored by Izhaki based on analysis of tens of millions of merchants. The research identifies repeatable scam templates, including synthetic merchant identities, deceptive storefront networks, scam subscription structures, and transaction laundering operations, arguing that these patterns behave like scalable systems that adapt faster than traditional detection methods.

Ballerine, founded in 2021, positions itself as an AI-native risk intelligence platform for financial institutions, acquirers, and payment providers. Its agents continuously analyse business registries, transaction data, and web intelligence to produce risk decisions across the full merchant lifecycle.

Why This Matters to FinanceX Readers

For acquirers, banks, and payment service providers, the economics of merchant fraud are shifting faster than compliance budgets. Every day a synthetic merchant operates undetected translates directly into chargebacks, scheme fines, and regulatory exposure under frameworks such as PSD3 and the FCA's Consumer Duty rules.

Investors evaluating fintech infrastructure plays should watch merchant underwriting as an increasingly strategic category: it sits at the intersection of AI, compliance, and payments, and the firms that can compress decision times from days to seconds will command pricing power as agentic commerce scales.

By Koen Vanderhoydonk - FinanceX Magazine