Google has announced plans to use machine learning technology to estimate the age of its users. The company is already testing this new model in the United States, with plans to gradually introduce it in other countries.
The machine learning model will analyze available user data, including browsing history, YouTube viewing habits, and account age, to assess whether the user is under 18. If a user is identified as a minor, Google will send a notification to change the settings and provide instructions on how to verify their age with a selfie, credit card, or government document.
The new age verification feature will work in conjunction with Google's existing security measures for minors. For such accounts, SafeSearch filters will be activated to block explicit content in search results, and access to YouTube videos deemed inappropriate for people under 18 will be restricted.
Google's decision to introduce age verification comes amidst increased attention to child online safety in the United States. Lawmakers are currently debating initiatives such as the Kids Online Safety Act (KOSA) and COPPA 2.0, as well as the Kids Online Social Media Protection Act (KOSMA), which aims to ban children under 13 from social media. KOSMA requires platforms to estimate the age of users using algorithms, although critics argue that this method is not always reliable.
It is worth mentioning that Meta, the parent company of Facebook and Instagram, also started using AI last year to detect "signals" that a user might be under 18.
In the coming weeks, Google will also introduce additional parental control features. Parents will be able to limit calls and messages on their children's devices while they are at school. Starting next month, parents will be able to add contacts to their children's phones through the Family Link app, limiting calls and messages to approved numbers. Google also plans to introduce features that will allow parents to add and manage payment cards in their children's Google Wallet.