Gartner: 30% of Enterprises Will Find Identity Verification Unreliable by 2026

In February 2024, Gartner published a prediction that should concern every organization relying on biometric or knowledge-based identity verification: by 2026, 30% of enterprises will no longer consider these solutions reliable when used in isolation, due to AI-generated deepfakes.

The implications are significant. Facial biometrics, document verification, voice authentication, and knowledge-based questions, the identity verification methods used by most organizations for account recovery, onboarding, and high-assurance scenarios, are all vulnerable to synthetic media generated by increasingly accessible AI tools.

The MGM Resorts breach demonstrated the social engineering dimension. But deepfakes add a technological dimension: an attacker can now generate convincing video, voice, and document forgeries that pass automated verification systems designed for human review.

This does not mean identity verification is worthless. It means that single-factor verification of any kind, including biometrics, is no longer sufficient for high-assurance identity decisions. Organizations will need multi-layered, contextual verification that combines device signals, behavioral analytics, network context, and multiple attestation methods.

For identity architecture, this prediction reinforces the need to design verification systems that are resilient to any single factor being compromised. No single authentication or verification method should be trusted in isolation. The architecture must be designed for the assumption that each individual factor can be defeated.