The Latest NIST FATE Benchmark Provides Independently Tested, Publicly Available Insights Into the State of the Industry, Helping Businesses Make More Informed Decisions
Denver, CO – ROC, the leading American-made multimodal biometrics provider, today announced its top-tier ranking in the recent NIST Face Analysis Technology Evaluation (FATE) for Age Estimation and Verification. The report underscores the significant advancements in age estimation technologies, highlighting their real-world potential as cutting-edge, privacy-conscious solutions for managing access to age-restricted products, services, and content.
In the most recent NIST FATE benchmark, ROC was the top U.S. provider of age verification technology, and second globally, with an average Mean Absolute Error (MAE) of 2.96 years across three main datasets – application, border, and mugshot.
“Once again, we are thrilled to be recognized through NIST benchmarking as a leading provider of highly accurate age estimation technology. As the top-ranked American-made biometrics company in this space, we take great pride in pushing the boundaries of Vision AI research and performance to meet the rigorous standards of our military, law enforcement, and global FinTech partners.”
– Dr. Brendan Klare, Chief Scientist and Co-Founder, ROC
A Privacy-First Approach to Age Estimation and Verification
ROC age estimation is a privacy-first technology that accurately verifies a user’s age without requiring their date of birth or other sensitive personal information. This helps businesses uphold stringent privacy standards while prioritizing user privacy, making it an ideal choice for industries where age-restricted content or services are prevalent, such as retail, social media, online gaming and gambling, and adult content platforms.
Border Challenge
ROC placed #2 overall in the Border Challenge-25, making it the top U.S. provider.
Values are averages of 12 FPR estimates from two sexes and six regions-of-birth for subjects aged 17. Comprises of airport concourse border crossing photos with variable pose, illumination, and eye glasses.
Child Online Safety
In a tight finish, ROC placed #2 overall in Child Online Safety for Ages 17-22, the key cohort for protecting minors.
FPR is for proportions of subjects aged 17 to 22 but whose age is estimated from 13 to 16 (below 17).
“ROC has made protecting the safety of children one of its core missions by ensuring age-appropriate access to online content and services. As regulatory bodies tighten restrictions and consumer awareness grows, we have seen a massive jump in demand for reliable and robust age estimation technologies. We take age estimation very seriously and see immense potential in the role of ethically developed solutions that safeguard minors while simultaneously protecting user privacy, empowering our clients to operate more securely, efficiently, and in full compliance.”
– Blake Moore, COO, ROC
Age Estimation: Part of the ROC Identity Toolkit
ROC age estimation is part of its comprehensive multimodal identity toolkit, which includes face recognition, fingerprint recognition, iris recognition, and next-gen ABIS capabilities. This plays nicely with the company’s other primary Vision AI discipline: live video analytics encompassing real-time threat detection, gun detection, and situational intelligence. These combined capabilities, available on mobile and edge devices, are trusted by organizations across diverse industries to enhance security, identify users, and comply with regulatory requirements.
Navigate Your AI Supply Chain with NIST Standards
NIST is the leading performance benchmark for age estimation, face recognition, and other biometric standards. The recent NIST FATE report provides businesses looking for trusted age estimation solutions with independently verified data to compare the performance and provenance of their technology providers.
“Our customers understand the importance of knowing their AI supply chain to ensure the integrity of their systems. By choosing ROC’s American-made, military-grade solutions, they inoculate themselves from the ever-present threat of poison AI.”
– Dr. Brendan Klare, Chief Scientist and Co-Founder, ROC
With an unwavering commitment to homegrown innovation and ethics, ROC high-fidelity algorithms, derived from massive ethically sourced datasets and fully designed and developed in the USA, are proven to deliver exceptional accuracy with zero bias. Similar to our other products, ROC age estimation is designed to be minimally intrusive and considerate of fast-evolving industry regulations and individual privacy rights.
Making the world smarter, safer, and stronger.
ROC is the only American-made multimodal biometrics and computer vision provider, trusted by the U.S. military, law enforcement, and leading global FinTech brands. Their fully multimodal solutions deliver game-changing results with battle-tested reliability across NIST government testing, military applications, and 200M+ annual identity proofing transactions. ROC is employee-owned, ethics-driven, and proudly made in America. The company is headquartered in Denver, CO, with centers of excellence in Morgantown, WV, and Grand Rapids, MI.
- Face recognition
- Fingerprint recognition
- Iris recognition
- AI video analytics & threat detection
- Vehicle & license plate recognition