Asking the FTC to approve facial age estimation for verifiable parental consent

profile picture Yoti 3 min read
Father and son using devices together

Together with the Privacy Certified program of the Entertainment Software Rating Board (ESRB) and Kids Web Services Ltd, a wholly owned Epic Games subsidiary, we are asking the Federal Trade Commission (FTC) for approval to implement facial age estimation as an authorised method for verifiable parental consent (VPC). The Children’s Online Privacy Protection Act (COPPA) requires companies to ensure they are not collecting personal data from children under the age of 13 without a parent’s consent. Currently, the COPPA Rule enumerates seven, non-exhaustive methods that enable parents to verify their consent. These include verification by government ID, credit card transaction, and a method that involves facial recognition, which is different from what we propose in our application.

We’d like the FTC to authorise facial age estimation as another VPC method to give parents and operators more choice in how parents can prove their age and grant consent. We are not seeking approval to implement this technology to check if children are old enough to purchase, download, and/or play a video game. Children are not involved in the age estimation process at all, which is designed to confirm that parents are adults, as required under COPPA.

We’d like to explain how the FTC application seeks to implement the technology in a way that is consistent with COPPA’s requirements for data minimisation, confidentiality and security.

  • Facial age estimation provides an accurate, reliable, accessible, fast, simple and privacy-preserving method to ensure that the person providing consent is an adult.
  • Facial age estimation is not facial recognition; it estimates a parent’s age without identifying them. It does so by converting the pixels of a facial image into numbers and comparing the pattern of numbers to patterns associated with known ages.
  • Facial age estimation does not create a database of faces. It doesn’t learn the user’s name, identity or anything about them. It does not scan their face against a database.
  • The technology is inclusive; it requires no collection of identity or payment card information.
  • It is accurate and does not show material bias among people of different skin tones. We have done extensive testing based on millions of facial scans and publish the accuracy levels transparently.
  • In our proposed application, facial age estimation is always presented as an option to parents alongside other approved methods of verification, providing the parent with a choice of methods.

If you have any questions about our FTC application, please get in touch.

Keep reading

Image of a young man looking down at his phone. Surrounding him are three illustrations: a shield representing security, a building representing financial services and a smartphone representing how a Digital ID app can be used for identity verification.

How Digital IDs are streamlining everyday tasks for millions

We’ve recently just hit a huge milestone: our Digital ID apps have had 14 million global downloads. Businesses and individuals are increasingly using Digital IDs. In fact, experts have predicted that the number of digital identity verification checks will surpass 70 billion by the end of 2024. This blog explores how our Digital IDs are making people’s lives easier and more secure.   But first, what is Digital ID? Instead of using a physical document, such as a passport or driving licence, a Digital ID is a way for people to prove who they are on their phone. Creating

10 min read
A man placing his face in the frame to perform a facial age estimation with Yoti

How accurate is facial age estimation?

“How accurate is it?” is the first question regulators, businesses and users tend to ask about facial age estimation. To date, we have mainly presented the technology’s Mean Absolute Error (MAE) as a proxy for accuracy. It’s an intuitive way to understand how accurate a model is. We can say it’s accurate to 1.3 MAE for those aged between 13 and 17 years or 2.5 MAE for those aged between 6 and 70 years. However, the answer is slightly more complicated. Following the COVID-19 pandemic, many people will be more aware of the terms ‘true positive’ and ‘false negative’

5 min read
An image of man with an 'over 18' facial age estimation credential. Around him are icons representing the gambling, gaming, financial services and retail industries.

How Yoti’s facial age estimation is used across different industries

Checking users’ ages has never been more critical for businesses catering to diverse audiences. However, they’re faced with the challenge of effectively verifying the ages of their users whilst maintaining seamless and user-friendly experiences.  Yoti’s facial age estimation is a secure, privacy-preserving way to do just that. Our technology is used across a variety of industries, both online and in-person. This includes retail, social media, dating, gaming, gambling and financial services. In this blog, we explore how businesses are using facial age estimation to create safer, more positive experiences for their users.   What is facial age estimation? Facial

10 min read