Ballerine Launches New Scam & Fraud Detection API

Ballerine, a merchant risk and compliance platform for acquirers and payment service providers (PSPs), introduced to market its new agentic detection solution, the Scam & Fraud Detection API, that identifies high-risk merchants and fraud networks in less than 60 seconds before transactions occur.

Historically, it has usually taken acquirers anywhere from days to weeks to sort out high-risk or fraudulent merchants because the compliance process relied on fragmented tools, onboarding snapshots and manual reviews. In addition to the long lag between initial onboarding and finding problematic merchants, risky merchants entered portfolios without being flagged by analysts at all because those conducting the review did not have the time, context, or investigative depth to connect the dots during onboarding.

Ballerine addresses both problems. It compresses detection into a pre-transaction window, with results delivered in under 60 seconds, and it automates the investigative layer by connecting signals, surfacing inconsistencies, and enabling confident decisions at onboarding and during ongoing monitoring.

“The industry is facing a structural gap as fraud has evolved into coordinated, AI-powered operations, while many risk systems are still built for a slower, manual world. If it takes days to determine whether a merchant is risky, that gap is already being exploited at scale,” said Noam Izhaki, Co-founder and CEO of Ballerine.

Instead of relying solely on transaction monitoring or static onboarding checks, the Scam & Fraud Detection API analyzes the broader context around a merchant, including digital footprint, behavioral signals, ownership structures, and ecosystem connections, ensuring that fraudulent activity is stopped in its tracks.  According to Izhaki, it builds and maintains a dynamic merchant profile across the merchant lifecycle and can enrich or reassess the profile in real-time when a new API call comes in.

“This is not just a static cache, and it is not just a one-off live query;  it is a continuously updated intelligence layer that can be acted on in real time,” he emphasized.

The solution uses this intelligence to make a contextual risk judgment with supporting evidence regarding the intent of the merchant in question. Contextual variables can include claimed business model, website behavior, transaction patterns, network relationships, digital footprint, policy language, and observed changes over time.

What is key is that the solution seeks to look across the full merchant context to find whether the facts are consistent with a legitimate merchant who is not high-risk or whether the merchant should be flagged as suspicious based on the supporting evidence.