Block Enhances Fraud Detection With Real-Time AI
Block Chief Risk Officer Brian Boates tells PYMNTS the company is shifting fraud defense to real-time machine learning that evaluates transactions using thousands of behavioral signals to intervene before money leaves accounts. The article highlights AI-enabled impersonation, exposed identity data, and the use of selfie/liveness checks, while describing selective "smart friction" to block high-risk transactions without broad user disruption.
Key Points
- 1Deploys real-time ML evaluating transactions with thousands of behavioral signals to block fraud
- 2Warns AI-enabled voice/chat impersonation and exposed identifiers degrade traditional identity verification
- 3Implements selective smart friction to pause high-risk payments without causing alert fatigue or churn
Scoring Rationale
Authoritative, industry-relevant coverage of real-time ML fraud defense; limited technical novelty and no new empirical benchmarks.
Sources
Public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
