Multiple posts on Reddit and X have alleged that Anthropic has incorrectly flagged users as under 18 and suspended their Claude accounts. In emails sent to affected users, Anthropic states: “Our team found signals that your account was used by a child. This breaks our rules, so we paused your access to Claude.” The company adds, “If you think we made a mistake and would like to turn your account back on, please use this link to verify your age,” noting the link will expire in 30 days.

One Reddit poster wrote: “The Anthropic Team just saw all of my conversations and locked me out.” The user added that they are on the Pro Plan and over 18 and are attempting to appeal the decision. They add that Anthropic is using Yoti for age verification via digital ID, facial scan, or biometrics. Comments state some accounts have been reinstated after submitting ID. Another said their subscription was refunded after being wrongly flagged, and “my projects are completely broken now”. 

Anthropic’s policy: In a December 2025 announcement, Anthropic said it requires all Claude users to be 18 or older and asks them to affirm their age at sign-up. It stated its classifiers flag conversations where users self-identify as under 18, triggering review and possible suspension, and that it is developing systems to detect subtler signals. Meanwhile, it has also said that app store age verification in some US states may prevent under-18 users from creating or signing into Claude on the mobile app.

Age verification and privacy: An open letter signed by over 400 scientists and researchers warns that age verification systems expand the collection of sensitive personal data, including biometrics, behavioural signals, and contextual information. It states these systems introduce risks of misuse by providers, third-party access, and data breaches. Further, the signatories say such tools can produce high error rates and biased outcomes, while enabling broader tracking and profiling as they link to legacy identity infrastructures. 

As adoption grows, they caution that privacy risks will scale across services. Notably, an October 2025 breach at Discord exposed around 70,000 images of government IDs submitted for age verification.

Questions for Anthropic: MediaNama has reached out to Anthropic with the following questions. The article will be updated when we receive a response.

• What specific signals or behavioural indicators do your classifiers rely on to flag accounts as potentially under 18, beyond explicit self-identification in conversations?
• What is the measured accuracy of these systems, including false positive rates, and how have you evaluated them across different user groups?
• What safeguards are in place to minimise wrongful suspensions, particularly for paying subscribers, and how often are such decisions overturned on appeal?
• When an account is flagged, what level of access do human reviewers have to user conversations and data during the review process?

Also read

• Andhra Pradesh explores DigiLocker age tokens for social media curbs on children aged 13-16
• Communications and Information Technology Committee calls for comprehensive AI law, age-based restrictions for social media platforms
• Australia extends age verification rules to porn sites, AI chatbots, and adult games