QWE AI Academy The real takeaway most people miss
Claude is only available to people over 18 years. Treat the signup checkbox like a joke and you can lose the account mid-project. Classifiers, Yoti reinstatement, and the July 2026 privacy language are live. Over-eager? Sometimes. Fixable? Yes – if you pick the path that keeps history and billing.
Already flagged, or trying not to join the Reddit pile-up? Skip the philosophy. Verify the flagged account the way that actually comes back.
Why lockouts spiked
Anthropic required 18+ (or a higher local consent age) from the start. You affirm it at signup. In their December 18, 2025 wellbeing post, they said younger users face higher risk from chatbot conversations, so classifiers watch for under-18 self-ID – and they’re building subtler-signal detectors. Hit those indicators and the consumer account gets disabled until age assurance passes.
Scale changed. Teeth too. The Age assurance help article (as of the current Claude Help Center text) spells out Yoti. Some US state rules push Apple/Google to share age ranges, which can block the mobile app even when you’re an adult. Privacy policy language effective July 8, 2026 covers age/identity checks for a small subset of flagged consumer accounts – gov ID image, selfie or video, facial geometry templates, verification result; Persona shows up in some identity flows per reporting on that update. Community threads (and Medianama-style coverage) are full of Pro users who logged in locked out.
Enough context.
Method A vs Method B when you hit the wall
Method A – wait it out or spin a fresh account. Sounds fast. New email, new checkbox. You often drag the same signals over, burn projects and custom instructions, and kiss paid entitlement goodbye. Users also report the verification link on the old account can expire in about 30 days – so stalling isn’t free.
Method B – finish official Yoti age assurance on the flagged account. Email link → strongest method your documents and camera support → pass/fail back to Anthropic. You keep history and billing. Data handling is simpler than the panic threads suggest.
B wins for almost anyone who still needs that workspace. A only makes sense if the account is disposable and you refuse any third-party check.
Yoti walkthrough that actually restores access
Flag email includes a verification link. Open it on a phone or laptop with a usable camera – not the worst webcam in the building.
Three options on the official age-assurance flow:
- Facial age estimation – selfie only, no document. Fast. Fails more often on young-looking adults, dim rooms, or soft laptop cams.
- ID verification – passport, driving licence, or national ID (Yoti’s country list). Clear photo, full document, no glare. This is the reliable fallback.
- Yoti Digital ID app – share an already-verified “over 18” attribute if you have one.
Have a physical ID and any doubt about the selfie? Start with ID verification. Don’t grind the same failing face shot.
If facial keeps returning “could not confirm,” change device or switch to ID immediately. Community reports line up with that: better phone camera or a straight document photo clears it. Don’t burn the link window.
Yoti finishes → Anthropic gets pass/fail only. Per Claude’s help docs, Yoti is SOC2 audited and deletes selfie, document images, and personal data right after the check; Anthropic never sees those images.
Pass lands, open billing and projects the same hour. Gifted or paid Pro stuck on Free while billing still looks active? Separate backend glitch – support ticket with verification proof and billing reference. Don’t open a second account to “fix” it; that usually makes reconciliation worse.
New accounts: affirm 18+ honestly, don’t write “I’m 17” for fun, skip underage roleplay cosplay, stay on a supported country/region.
Edge cases that catch adults
Web works, mobile app dead? Often Apple or Google sharing an under-18 range from Family Sharing / Family Link. Anthropic can’t override store age data. Device Settings → Privacy / Family, stop sharing age range with apps, or contact Apple/Google Support with proof. Their minimum age access restriction article is blunt on this.
Old chats still bite. One throwaway “I’m 15” bit, heavy kids-game help (Roblox cheat sheets and friends), or a juvenile homework tone can surface months later. Classifiers keep history.
Fuller identity checks (Persona in some cases under the July 8, 2026 privacy language) may want live selfie plus physical gov ID. Photocopies, screenshots, digital-only IDs, temporary paper – usually fail. Still framed as a small flagged subset, not every login.
Docs still won’t say exactly which “certain circumstances” beyond age signals flip you into the heavier identity path. The policy wording is broad; coverage noticed.
Actually, the privacy noise is louder than the day-to-day friction for people who complete the age step once and move on.
FAQ
Can I still use Claude if I’m under 18?
No. Consumer Claude is 18+. No kid mode. No parental dashboard. Some families put a parent on the account; the holder is still the adult under the terms.
I verified with Yoti and my Pro plan disappeared – what now?
Known post-reinstatement glitch: billing side still shows the gift or sub, UI drops to Free. Ticket with the Yoti success mail plus invoice/order ID. Second account makes it harder, not easier.
Is giving Yoti my face or ID actually safe?
Help docs: delete-after-check on images/data; Anthropic stores pass/fail for that flow, not the biometrics. SOC2 on Yoti. Persona-style identity checks follow the newer privacy terms if you get pulled into that smaller bucket. Uncomfortable with any of it? Consumer product isn’t available another way – API/Team/Enterprise sit under different terms and aren’t the center of these consumer age locks.
Have the email? Open it. Prefer ID over selfie when anything looks iffy. Confirm projects and plan the minute access returns.