NIST Released New Facial Recognition Report

The latest evaluation measured accuracy and demographic disparities across dozens of software submissions.

Updated on Sept. 21, 2026 in Artificial Intelligence

Isometric editorial illustration of a geometric lattice of crystalline modular cubes, representing the systemic structure of algorithmic testing. — The National Institute of Standards and Technology released its latest facial recognition evaluation report, detailing accuracy benchmarks and demographic consistency across 21 software submissions. AI Illustration. Upload story photo >

Live Poll

Can facial recognition technology be trusted to perform accurately and fairly for all demographic groups?

On September 1, 2026, the National Institute of Standards and Technology (NIST) published its latest Face Recognition Technology Evaluation 1:1 report. The findings highlight performance rankings and persistent demographic inconsistencies among facial recognition algorithms.

Why it matters

This evaluation provides organizations with critical performance data to assess how different algorithms behave under specific deployment scenarios. It also serves as a benchmark for understanding how current technology handles demographic variety.

Cloudwalk Moontime v6 and Recognito v0 lead Visa-to-Visa matching with a 0.06 percent error rate. Meanwhile, 97.3 percent of algorithms demonstrate the highest disparity for West African subjects, while 98.2 percent show the most error for elderly users.

The players

NIST

The National Institute of Standards and Technology is a federal agency that promotes innovation and industrial competitiveness by advancing measurement science and standards.

Cloudwalk

Cloudwalk is a technology developer that produces facial recognition software and algorithms used in biometric security.

Innovatrics

Innovatrics is a provider of biometric software solutions specializing in fingerprint and facial recognition identification technology.

The details

Twenty-one developers submitted new or updated software for this cycle, allowing NIST to test accuracy in categories like Mugshot-to-Mugshot and Kiosk-to-Border. The report uses the Bias Gini score, which had a median of 0.62, to track demographic consistency across tested systems.

NIST released the report on September 1, 2026.

The Tech Race

This reporting tracks the iterative improvements in biometric accuracy against the backdrop of NIST's Face Recognition Technology Evaluation (FRTE) program. It underscores how the industry struggles to standardize performance while scaling recognition capabilities across diverse populations.

The persistent demographic disparities documented in the report suggest that biometric systems used in airports or government kiosks may not perform with equal reliability for all individuals. Users should remain aware that technical accuracy varies significantly based on age, gender, and ethnicity.

The takeaway

The latest NIST report demonstrates that while some algorithms achieve near-perfect matching accuracy, most systems still struggle with significant demographic bias. Developers and public agencies must prioritize these consistency gaps to ensure equitable technology deployment.