- Anthropic's consumer terms require Claude users to be at least 18, or older if local law sets a higher age to consent. On December 18, 2025, Anthropic said its classifiers flag chats where users say they're under 18 and that it disables accounts confirmed to belong to minors.
- If Anthropic's systems flag an account, the owner gets an email with a link to verify their age through Yoti, using a selfie, an ID document or the Yoti app. Anthropic says it receives only a pass or fail result. On mobile in some US states, a flag can instead come from app store age data, which only Apple or Google can correct.
- In April 2026, MediaNama and Digit reported adults saying they had been wrongly flagged. Users reported a 30-day window to verify. Anthropic didn't answer MediaNama's questions. Its age-check help articles were last updated in March and May 2026.
Contents
Why is Claude 18+? Because Anthropic's terms say so. Its consumer terms require everyone who uses the Claude apps to be at least 18, and Anthropic actively looks for accounts that belong to minors and disables them. That sets Claude apart from ChatGPT, which allows teenagers with a parent's permission. Some adults have also found their accounts locked after being mistaken for minors.
This explainer is based on Anthropic's consumer terms, privacy policy, help center articles and December 2025 announcement, plus reporting from MediaNama, Digit and TechCrunch, checked October 16, 2026. The steps below are from Anthropic's help center.
The short version
- The rule: Anthropic's consumer terms say you "must be at least 18 years old or the minimum age required to consent to use the Services in your location."
- How flags happen: if you tell Claude you're under 18, its classifiers flag the chat for review. Anthropic also runs systems that look for subtler signs of a minor. On mobile, in some US states, app store age data can block you.
- What happens: Anthropic says it disables accounts "based on indicators of minor activity" and asks the owner to verify their age before continuing.
- How to get back in: use the verification link in Anthropic's email to check your age with Yoti, by selfie, ID document or the Yoti Digital ID app. If the flag came from your App Store or Google Play age, only Apple or Google can fix it.
- What's new: no age-rule change in October 2026 in Anthropic's terms, help center or policies, checked October 16. The current privacy policy, effective September 10, 2026, describes age and identity checks.
Why does Claude require users to be 18?
Anthropic's consumer terms set the minimum at 18, or higher where local law requires. Its privacy policy, effective September 10, 2026, says the services "are not directed towards" children under 18 and that Anthropic has "measures in place to detect and remove children from our Services."
Anthropic explained its enforcement, but not a detailed rationale, in a December 18, 2025 post on user well-being. It said "we require Claude.ai users to be 18+ to use our product," and that all users "must affirm that they are 18 or over while setting up an account." It described age enforcement as part of its user-safety work, alongside handling conversations about suicide and self-harm, and said it had joined the Family Online Safety Institute.
Other AI apps handle minors differently:
| App | Minimum age | What happens if you're judged to be under 18 |
|---|---|---|
| Claude (consumer) | 18 | Account can be disabled until you pass an age check |
| ChatGPT | 13; under 18 needs a parent's or guardian's permission | Account moves to "ChatGPT for Teens," with extra safeguards; adults can verify with Persona |
| Character.AI | Open-ended chat ended for under-18 users by November 25, 2025 | Moved to a separate under-18 experience |
Last verified: October 16, 2026. Sources: Anthropic consumer terms, OpenAI Terms of Use, OpenAI age prediction help article, Character.AI announcement.
The 18+ rule covers the Claude apps. Developers using Anthropic's API can build products for minors if they follow Anthropic's guidelines for organizations serving minors. These require age verification, content moderation, compliance with laws such as COPPA, and telling users they're talking to an AI. Anthropic says it periodically audits these organizations.
How does Claude decide an account belongs to a minor?
There are three documented routes.
- You say so. Per Anthropic's December 2025 post, if a user identifies as under 18 in a conversation, "our classifiers will flag this for review," and Anthropic will disable accounts "confirmed to belong to minors."
- Subtler signals. The same post said Anthropic was building a classifier for "other, more subtle conversational signs" that a user may be underage. The help center's "Age assurance on Claude" article, last updated May 18, 2026, says Anthropic has "safety systems in place to detect if people under 18 may be using Claude" and will "disable accounts based on indicators of minor activity." Anthropic hasn't published which signals it uses or how accurate its systems are.
- Your app store says you're under 18. The help center article "Minimum age requirement access restriction", last updated March 16, 2026, covers the mobile app. New laws in certain US states require app stores to verify users' ages and share that information with developers. If the App Store or Google Play marks you as under 18, the Claude app blocks you. Anthropic says the rollout is staged state by state.
What happens when Claude flags your account?
Per the age assurance article, a user whose account shows signs of belonging to a minor must verify their age before they can continue. The verification link comes in the notification email, and passing the check reinstates the account.
In April 2026, adults began reporting these emails publicly. MediaNama (April 15) and Digit (April 12) reported Reddit and X posts from users who said they were over 18, including a Pro subscriber, whose accounts were paused. The emails reportedly said Anthropic had "found signals that your account was used by a child," and users said the verification link was valid for 30 days. Commenters said some accounts were restored after an ID check. One said their subscription was refunded, and another said their projects were broken.
Two points from those reports remain unconfirmed. One user claimed staff had read their conversations. MediaNama asked Anthropic what access reviewers have, and Anthropic hadn't responded by publication. Anthropic's December post says self-identified minors are flagged "for review," but doesn't say who reviews or what they see. The 30-day window comes from user-shared emails, not from a help center page.
How do I unlock a Claude account flagged as a minor?
These are the documented steps, from the two help center articles as checked October 16, 2026.
If you got an email saying your account was flagged:
- Open the verification link in the email. Per the reports above, it may expire, so don't wait.
- Choose one of the three methods Yoti offers:
- Facial age estimation from a selfie, with no ID needed
- ID document upload, such as a passport, driver's license or national ID card (Yoti lists country-specific options)
- Sharing a verified "over 18" attribute from the Yoti Digital ID app
- If you pass, Anthropic reinstates the account.
If the mobile app says you don't meet the minimum age:
- Check the age on your Apple Account or Google Account. Anthropic says it can't see or change your app store information.
- If you're in Apple Family Sharing or Google Family Link, a parent or guardian may have set your age. You may need them to update it.
- If the age is wrong and you can't change it, contact Apple Support or Google Play Support.
What Anthropic sees. Per the age assurance article, Anthropic doesn't see your ID or image and receives only a pass or fail result. It says Yoti deletes selfies, document images and personal data once the check is complete, and that Yoti is independently audited for SOC 2 compliance.
Is this the same as Claude's identity verification?
No. Anthropic runs a separate identity verification process through a different company, Persona. That help article, last updated August 11, 2026, says you may be asked to verify "when accessing certain capabilities," during "routine platform integrity checks" or for other safety and compliance reasons. It requires a physical government photo ID and sometimes a live selfie. Digital IDs and student IDs aren't accepted.
When Anthropic added these checks to its privacy policy in June 2026, it pointed TechCrunch to a post by its Thariq Shihipar saying the change was "unrelated to the Fable or Mythos rollout." Shihipar described it as an update to the appeals process for accounts flagged for possible fraud. For why some models aren't available to everyone, see our explainer on why Claude Fable 5 is unavailable.
If your account was banned for something other than age, such as Usage Policy violations, the identity verification article points to a separate appeal form reviewed by Anthropic's Safeguards team. Our explainer on Anthropic's usage policy and how it's enforced covers what can lead to a ban.
Has anything changed recently?
Not as of October 16, 2026, according to Anthropic's terms, help center and policies:
- Help center: the age assurance article was last updated May 18, 2026, and the app store article on March 16, 2026.
- Privacy policy: the current version took effect September 10, 2026. It describes verification data that may include an ID image and details, a photo or video, "facial geometry templates," and the result of the check. It doesn't name a provider except for Korean users.
- Usage policy: Anthropic's October 8, 2026 update covered abusive behavior toward Claude and other areas, not the age rule.
What's still unknown: Anthropic hasn't published false-positive rates, which signals trigger a flag, how many accounts have been disabled, or what happens to a flagged account's subscription and data if the owner never verifies. Our Claude app explainer covers the plans, and the age rule applies to all of them.
What's confirmed, reported and uncertain
- Confirmed (Anthropic documents): the 18+ rule; self-identification flags and account disabling; signal-based detection; Yoti's three methods and pass/fail-only reporting; app store age enforcement in some US states; Persona-based identity verification; the September 10, 2026 privacy policy.
- Reported: wrongful flags of adults (MediaNama, Digit, April 2026, based on user posts); the 30-day verification window; refunds and broken projects (user comments); Shihipar's statement that identity checks were unrelated to Fable or Mythos (TechCrunch, June 22, 2026).
- Uncertain: which signals trigger flags; how accurate they are; whether humans read flagged conversations; what happens to accounts that never verify.
About this storyBased on the sources linked below. Editorial standards
