Financial fraud is increasingly complex, spreading across networks of accounts, wallets, devices, and transactions. It rarely appears as a single event, but through relationships and evolving behavior that traditional analytics often miss.

In this session, you’ll learn:

• Why fraud detection is a graph problem
• How graph transformer models uncover coordinated fraud activity
• How Coinbase combines transaction, device, and behavioral data to detect fraud at scale

You’ll leave with a clear understanding of how graph-based AI and embeddings power modern fraud and risk systems across financial services and insurance.