Discord Shifts to Automatic Age Checks, With ID Scans as Backup
Discord has begun rolling out its revised age assurance system, which places users in Adult or Teen groups automatically instead of asking for documents first. The rollout starts this week, according to reporting published Sept. 22. Engadget
The sorting is inference-based, meaning Discord estimates an age group from account clues rather than checking ID. It looks at markers such as account age, servers joined and activity levels. It does not read messages or listen to calls to make that estimate. It is closer to sorting by the outside of the envelope than opening the letter. The assigned group does not show on a profile and is not shared with other server members. Users can check their assignment under Account Status in User Settings. Discord expects 90 percent of users will need no further action if placed in the correct group.
That approach replaces an earlier plan that would have required facial age estimation, a software estimate of age from a face scan, or an ID scan for users not confirmed as adults by internal tools. Discord announced in February that it would delay those changes in response to privacy and personal data collection concerns.
The February response was led in part by the chief technology officer, who addressed community concerns in a company blog post. The post is listed under Policy & Safety as "Getting Global Age Assurance Right: What We Got Wrong and What's Changing". In it, Discord stated there will be no mass ID collection for age assurance and announced new vendor transparency commitments, or clearer public information about the outside firms it uses. Discord
Timing shifted as well. Discord delayed the global rollout of age assurance until the second half of 2026 to expand verification options. Discord An August 2025 policy update had described a privacy-forward age verification experience, one designed to collect less personal data, required in specific scenarios that meet OSA requirements. Discord
Accounts the system cannot place fall into an Unconfirmed category. Those users can verify they are older than 18 via credit card, Apple App Store, Google Play Store, Google Wallet, AgeKey credential service, ID scan, or video selfie, depending on regional restrictions. UK users can confirm their age group by using a Passport ID already verified and stored in Google Wallet. Discord Support
Data handling constraints are explicit. Information submitted for age verification is stored only for the minimum time necessary and in most cases is deleted immediately. Discord will require any third-party age assurance vendor to permanently delete user information once verification is completed. Any facial age estimation must run completely on the user's device, so the face scan is processed locally rather than uploaded.
Teen accounts receive restrictive defaults. Messages from non-friends go to the message requests inbox. Teen accounts receive an alert before accepting a friend request from a user without mutual friends or shared membership in a small server. They cannot view sensitive-tagged images or adult-only servers, channels and bot commands. Full profile details and activity are accessible only to friends and members of smaller servers.
Accounts in the Unconfirmed category are placed under the first two Teen restrictions until the user confirms age. That means message-request routing and friend-request alerts apply by default.
The broader context here is a shift from document-first verification to exception-based assurance. Most users are classified passively from account metadata, while only edge cases touch payment rails, platform attestations like app store checks, reusable credentials like AgeKey or Google Wallet, or biometric estimation from a face. For practitioners, the interesting controls are architectural: no content inspection, no public age-group signal, on-device estimation, and contractual deletion obligations for vendors. In my view, that combination is more durable than the original scan-everyone approach. It reduces data collection while preserving a path to stricter checks where regulation or safety policy requires it.
