If you only want to watch videos intended for adults, be prepared to show your ID (or a credit card if you’re in the US).
Back in February, YouTube warned us that the AI feature that determines viewers’ ages and displays appropriate content is coming. According to a YouTube blog post, this machine learning–based content filtering and user account classification method has been used for some time in a few unidentified markets. However, in the coming weeks, YouTube will begin testing the technology on a small group of users in the United States to estimate their age. Teenagers will be treated as teenagers, and adults will be treated as adults.
The site will closely monitor the technology before rolling it out more widely. According to the video-sharing platform, the system estimates users’ ages and then provides an experience and protection appropriate to their age using this information, regardless of the date of birth provided in their account. For example, if YouTube decides that a teenager is watching a video, then personalized ads may be disabled, digital wellness tools may be enabled, and safety measures may be applied to recommended videos. This may also include restricting the repeated viewing of certain types of content. What does digital wellness entail?
The system interprets various signals to determine whether the user is over or under 18 years of age. These signals include the types of videos the user searches for, the categories of videos the user views, and how long the account has been active. If the system incorrectly estimates that a user is under 18, the user can verify their age with an ID or credit card. Only users presumed or verified to be over 18 can view age-restricted content not suitable for younger users.
It will be interesting to see how accurate the AI age verification system is. AI will undoubtedly cause as many problems for YouTube and its viewers as it solves. The February blog post that first mentioned using machine learning to filter user accounts emphasized the benefits of AI. A lot of poor-quality AI content is uploaded to YouTube every day, and the company is currently either unwilling or unable to filter it all out.
It remains to be seen whether AI will be a savior or a curse for the platform—or for us.
