Visa is partnering with Tonic.ai to produce synthetic transaction datasets that reflect real payment and fraud patterns—aimed at letting merchants and fintechs test and build without touching sensitive cardholder data.
Visa partners with Tonic.ai to synthesize transaction data for merchant and fintech testing
Visa announced a partnership with Tonic.ai focused on generating synthetic transaction data that mirrors payment and fraud patterns. The stated goal is to enable secure testing environments so merchants and fintechs can develop and validate payment experiences without using sensitive cardholder data.
In practice, this positions synthetic data as a safer stand-in for production payment data in development and QA workflows—especially when teams need realistic behavior (including fraud-like signals) but want to avoid the operational risk and compliance overhead that comes with copying real transaction records into non-production systems.
- Fraud-model iteration without raw PAN exposure: Data science teams can test fraud detection logic against realistic patterns while reducing the chance that sensitive cardholder data ends up in dev, staging, or vendor sandboxes.
- Potential PCI-DSS scope reduction: If non-production environments can rely on synthetic datasets rather than real cardholder data, security teams may be able to shrink the number of systems that fall under PCI-DSS controls (subject to internal interpretation and assessor guidance).
- Faster partner integration testing: Merchants and fintech partners often need data that “behaves” like production to validate payment flows; synthetic datasets can improve test fidelity without requiring production extracts or complicated masking pipelines.
- Privacy-by-design alignment: Replacing raw transaction data in routine development workflows supports GDPR-aligned engineering practices by minimizing use of personal data where it isn’t strictly required.