Yoti Makes Major Advances in Facial Age Estimation Accuracy Before OSA Dead

Published Date : 7/22/2025 

Yoti, a leading provider of facial age estimation technology, has announced significant improvements in the accuracy of its age estimation methods, just days before the UK's Online Safety Act (OSA) takes effect. The company's latest white paper reveals a mean error rate (MEA) of 1.1 years for estimating the ages of individuals between 13 and 17 years old, a crucial age range for social media platforms. 

Yoti, a leading provider of facial age estimation technology, has announced significant improvements in the accuracy of its age estimation methods, just days before the UK's Online Safety Act (OSA) takes effect. The company's latest white paper reveals a mean error rate (MEA) of 1.1 years for estimating the ages of individuals between 13 and 17 years old, a crucial age range for social media platforms.

The white paper, published days ahead of the OSA deadline, highlights the importance of highly effective age assurance methods. The new responsibilities under the OSA will take effect on July 25, requiring digital platforms to implement robust age checks, particularly for those hosting adult content. This includes pornography platforms, peer-to-peer sites, search engines, instant messaging services, dating apps, and others.

Yoti's facial age estimation technology has achieved a mean error rate (MEA) of 1.1 years for individuals aged 13 to 17, which is particularly significant for the 15- and 16-year-old age limits for social media. For individuals aged 18 to 24, Yoti’s MAE is now 2.1 years. The testing also shows that Yoti’s technology will correctly estimate 99.3 percent of youths between 13 and 17 as being under 21, and 99 percent of children between 6 and 12 as being under 13. The variance based on skin tone for TPRs of 13-17 year-olds across skin tones 1, 2, and 3 was 0.7 percent, which the company describes as “no discernable bias.”

MEA for people aged 6 to 70 years old is 2.4 years. Yoti CEO Robin Tombs emphasizes the growing demand for effective age assurance and increasing regulatory engagement. “We have continued to invest and improve the accuracy of facial age estimation. It is fast becoming an essential tool to help protect young people online, create safer, age-appropriate experiences, and respect people’s privacy. The improved accuracy means even more people will now be able to use our privacy-preserving facial age estimation to easily prove their age online,” Tombs stated.

The white paper also provides guidance on safety thresholds, buffers, and the “waterfall approach,” explaining evaluations of facial age estimation technology by NIST and ACCS and offering lessons from experience. Online age assurance checks in the UK are about to spike, with platforms required to verify whether their users are children or not if they host any adult material. This includes a wide range of platforms, from pornography sites to dating apps and instant messaging services.

“We are seeing a significant shift in how adults choose to prove their age. The majority opt for our facial age estimation to quickly and easily verify their age. For those estimated below the threshold, they can use our privacy-preserving Digital ID app or share their age from an official ID document,” Tombs added. As age checks become more common, nearly 1 in 4 people are using a reusable age token or Digital ID to anonymously prove their age. Yoti is proud to be protecting children online and introducing people to privacy-first age solutions, ensuring that online safety does not come at the cost of privacy. 

Frequently Asked Questions (FAQS): 

Q: What is the Online Safety Act (OSA)?

A: The Online Safety Act (OSA) is a UK legislation that requires digital platforms to implement robust age verification methods to ensure that users, especially children, are protected from harmful content. It takes effect on July 25, 2025.

Q: What is Yoti's mean error rate (MEA) for estimating the ages of individuals between 13 and 17 years old?

A: Yoti's mean error rate (MEA) for estimating the ages of individuals between 13 and 17 years old is 1.1 years.

Q: How does Yoti's technology perform for 18-to-24 year-olds?

A: For individuals aged 18 to 24, Yoti’s mean error rate (MEA) is 2.1 years.

Q: What is the 'waterfall approach' mentioned in the white paper?

A: The 'waterfall approach' is a method described in the white paper that provides guidance on safety thresholds, buffers, and the sequential steps to ensure effective age verification and estimation.

Q: How does Yoti ensure privacy while verifying age?

A: Yoti uses privacy-preserving methods such as facial age estimation, reusable age tokens, and Digital ID apps to verify age without compromising user privacy. 

More Related Topics :