Astra
Back to Legal Center
Safety

Safety & Age Policy

Minor protection and crisis intervention

Last updated · 2026-05-14

These terms are updated periodically; we email all users before material changes. This document is not legal advice for your specific situation — for that, consult your own counsel.

Safety & Age Policy

Last updated: 2026-05-14 · These terms are reviewed by Astra periodically and effective on the Last updated date above.

This policy describes the age tiers we enforce on Astra, the protections in place for minors, and our response to safety incidents. Read alongside the Community Guidelines, Crisis & Safety, Privacy Policy, and Terms of Service.

1. Age tiers

TierStatusWhat's available
Under 13Not permittedNone. Age gate blocks entry; we delete any account discovered to be under 13 (COPPA + global child-safety norms).
13 to 15 in GDPR-K 16-min jurisdictionsVerifiable parental consent requiredSame as 13-17 below, plus parental consent confirmation on file.
13 to 17Explore-onlyRead public content, watch public live streams, follow brands, run AI divination tools that do not involve voice or real-time exchange. Blocked: direct messaging with strangers, voice readings, live broadcasting, marketplace bookings, gifts, paid subscriptions outside parent-managed payment.
18+Full experienceEverything.

2. Verification

  • Self-attestation at the Cosmos / onboarding age gate is the floor — by accessing the Platform you represent your age.
  • For monetised features (gifts, payouts, live broadcasting, reader bookings) we require government-ID verification via Stripe Identity before activation.
  • We may re-verify when behaviour, content, or external reports suggest misrepresentation.
  • If we discover an account is under 13 we delete it immediately and notify the registrant.

3. What we do to keep people safe

3.1 Reporting + blocking

Block and Report are one tap from every star, message, stream, comment, profile, and reading. Reports go to Astra Trust & Safety with a target acknowledgement of 24 hours and a target decision within 5 business days (faster for live-safety matters).

3.2 Real-time content classification

We run automated classifiers across chat, captions, uploads, and voice for: CSAM, threats of violence, self-harm intent, hate speech, non-consensual intimate imagery, spam, scams, and identity fraud. Suspicious items go to human review, not auto-action.

3.3 Crisis routing

AI voice and chat trigger crisis-support scripting when high-risk language is detected. Hotline numbers appear in a non-dismissible card. Reader chats with crisis indicators pause for a safety check. See the Astrology & Divination Disclaimer § 6 and the Crisis & Safety document.

3.4 CSAM

Suspected child sexual abuse material is preserved, reported to the US National Center for Missing & Exploited Children (NCMEC) CyberTipline, and (where applicable) equivalent agencies in the user's country. Accounts that upload CSAM are permanently banned and reported regardless of the uploader's age.

3.5 NCII (non-consensual intimate imagery)

Removed within 24 hours of a verified report. No public ID is required to submit. Cooperation with StopNCII.org hash registry in progress.

3.6 Doxxing, swatting, stalking

  • Doxxing: same 24-hour removal standard.
  • Swatting threats: reported to local law enforcement where we have jurisdiction info.
  • Stalking: court orders honoured (see § 5 below).

4. Asking for help

ActionWhat happens
BlockRemoves the user from your feed, blocks DMs both ways, hides their content from you.
MuteHides their content from you without notifying them.
ReportHuman review, 24h target SLA. We share the outcome where lawfully possible.
Appeal a moderation actionSubmit at /help/appeals/account or /help/appeals/live within 30 days; decision within 5 business days.
Restraining-order coordinationEmail hello@astraplatform.ai with the court order; we restrict the named account from contacting you.

5. If you are a parent or guardian

  • We do not direct the Platform to children under 13. If you believe your child under 13 has an account, please email hello@astraplatform.ai with proof of guardianship; we delete the account.
  • For teen users in GDPR-K minimum-16 jurisdictions, your verifiable consent is required for the Platform to process personal data.
  • We provide a family-mode toggle (in development) that limits a teen's account to Explore-only content even where local law would permit more.

6. Crisis & emergencies

Astra is not a crisis service. If you or someone else is in immediate danger, call your local emergency number.

Always-available hotlines (full list in Crisis & Safety):

  • US: 988 · UK: 116 123 · AU: 13 11 14 · CA: 988
  • Mainland China: 400-161-9995 · HK: 2896 0000 · TW: 1925
  • DE: 0800 111 0 111 · FR: 3114 · BR: 188 · IN: AASRA +91-9820466726
  • Global: https://findahelpline.com

7. Vulnerable users

We make special accommodations for users who self-identify as in recovery, post-bereavement, or otherwise vulnerable:

  • Slow-reveal mode in Whispers and live chat
  • Safety nets that surface hotlines if conversation pattern matches risk indicators
  • Self-imposed limits on session count or spending in Settings → Wellbeing
  • We honour third-party requests from court-appointed guardians and conservators with valid documentation.

8. Banned content (one-strike)

The following are zero-tolerance and result in immediate permanent ban + NCMEC report (where applicable) + preservation of evidence for law-enforcement cooperation:

  • CSAM in any form
  • Real threats of violence
  • Glorification or planning of mass harm
  • Sale of weapons, controlled substances, fraudulent documents
  • Trafficking-related content

9. Three-strike for community violations

Less severe violations follow a graduated path: warning → 7-day suspension → 30-day suspension → permanent ban. See Community Guidelines § Enforcement for detail. Appeals are available at each step.

10. Contact

These terms are reviewed by Astra periodically and effective on the Last updated date above. For an active emergency call your local emergency number first.

Questions about this document?

Email us — we reply within 24h (12h for Pro & creators).