YouTube is launching a new AI-powered system for U.S. users that will estimate a person’s age and apply restrictions based on that determination. The “age estimation model” rolled out Wednesday to a small group of users, aiming to protect children from harmful content. Unlike older methods, it won’t rely on the birth date listed on the account but will use other signals to determine a user’s age, according to the Associated Press.
If the system makes an error, users will be able to override it. They can verify their age with a government ID, selfie, or credit card. The feature will only work for users who are logged in.
James Beser, YouTube’s senior director of product management for youth products, said the change ensures “teens are treated as teens and adults as adults.” He explained, “This technology will allow us to infer a user’s age and then use that signal, regardless of the birthday in the account, to deliver our age-appropriate product experiences and protection.” Beser noted the method has already been used in other countries “for some time” with good results.
Protections will mirror those already in place for accounts that indicate a user is a minor. That includes showing only non-personalized ads, turning on “digital wellbeing” tools like “take a break” and bedtime reminders by default, and displaying privacy prompts when uploading a video or commenting publicly. It will also reduce recommendations for videos that could be “problematic if viewed in repetition” and block access to age-restricted videos that are not meant for minors.