NIST Releases Guidelines to Detect Face Photo Morphs and Deter Identity Fraud

USA: NIST Releases Guidelines to Detect Face Photo Morphs and Deter Identity Fraud

Organizations in the United States can now consult new guidance released by the National Institute of Standards and Technology (NIST) on August 18, 2025 to help identify synthetic face images that could be used for identity fraud. The document, titled *Face Analysis Technology Evaluation (FATE) MORPH 4B: Considerations for Implementing Morph Detection in Operations* (NISTIR 8584), aims to assist agencies and private entities in deploying detection tools before morph attacks succeed.

Understanding Morphing Threats

Face‑morphing software blends the facial features of two individuals into a single image, allowing a malicious actor to present the composite as a legitimate photograph. When such a morph is submitted to facial‑recognition systems at borders, airports, or secure facilities, the system may erroneously match both original identities, enabling one person to assume another’s credentials.

Key Recommendations in the New Guidelines

The NIST publication provides a lay‑language overview of morph attacks and outlines practical steps for organizations that might encounter suspect images, such as passport‑issuing offices or border‑crossing checkpoints. It also advises on post‑detection actions, including escalation procedures and documentation.

Detection Performance Metrics

According to NIST computer scientist Mei Ngan, modern single‑image morph detectors can achieve up to 100% detection when trained on the specific morph‑generation software, while maintaining a false‑detection rate of 1%. However, performance may drop below 40% for morphs created with unfamiliar tools. Differential detection—comparing a questioned photo with a known genuine image—shows more consistent accuracy, ranging from 72% to 90% across both open‑source and proprietary morphing applications.

Operational Best Practices

The guidelines recommend a layered review process that combines automated detection, expert human analysis, and a defined workflow for handling flagged images. Ngan emphasizes that “the best procedures for investigating a candidate morph involve a combination of human review, the use of automated tools, and implementation of a process to handle review of suspected morphs.”

Preventive Measures

To minimize the risk of morphs entering operational systems, the report suggests controls during document application and issuance, such as capturing live biometric data and verifying photos against trusted sources before acceptance.

Overall, NIST’s effort seeks to raise awareness of morphing attacks and provide actionable guidance so that agencies can both detect and deter this emerging form of identity fraud.

This report is based on information from NIST, licensed under Public Domain (U.S. Government Work). Source: Official U.S. Government release.