New generative AI models are being used by fraudsters to generate convincing ID verification sessions at scale, making defense-in-depth more crucial than ever. To counter this, Plaid has introduced powerful new security features that make it harder for fraudsters to manipulate or bypass identity verification while streamlining the experience for trusted users. These enhancements include advanced machine learning models to detect deepfake biometric and synthetic ID detection, facial duplicate detection, age estimation, and new ML-powered user flow options such as risk-based escalations, selfie re-authentication, and trust index risk check.