What AI builders can learn from fraud models that run in 300 milliseconds

Fraud protection is a race against scale. For instance, Mastercard’s network processes roughly 160 billion transactions a year, and experiences surges of 70,000 transactions a second during peak periods (like the December holiday rush). Finding the fraudulent purchases among those — without chasing false alarms — is an incredible task, which is why fraudsters have been able to game the system. But now, sophisticated AI models can probe down to individual transactions,