Effective May 7, 2026 · Last updated May 7, 2026

App Policy & Community Guidelines

Bloom is a small social sanctuary. These rules describe how we expect people to behave inside the app, what kind of content is allowed, and how moderation works. They apply to every persona, sanctuary, voice room, story, clip, and direct message. By using Bloom you agree to these rules in addition to our Terms of Service and Privacy Policy.

1. Our core principles

Softer by default. We choose calm over loud, depth over reach, real over performative.
Safety is non-negotiable. We protect users — especially minors and vulnerable people — even at the cost of growth.
Personas, not anonymity. Personas let you express different sides; they do not let you escape accountability.
No dark patterns. No infinite scroll traps, no fake notifications, no public follower counts.
Human review. Real, trained reviewers handle escalations. Automated systems alone never make permanent bans.

2. Account & persona rules

One person, one Bloom account. Up to 4 personas per account.
Minimum age 16 (13 in the US, where lawful, with COPPA-compliant safeguards).
No impersonating real people, brands, public figures or other Bloom users.
No buying, selling, renting, or transferring accounts or personas.
No coordinated inauthentic behaviour (sockpuppets, brigading, follow-rings).

3. Content that is not allowed

The following content is removed on sight and may result in immediate suspension:

3.1 Child safety

Any sexual content involving minors. Reported to NCMEC and equivalent authorities, always.
Grooming behaviour, including attempts to move minors to other platforms.
Sexualised AI/illustrated depictions of minors.

3.2 Violence & dangerous behaviour

Threats, glorification, or incitement of violence against people or groups.
Content promoting terrorism, organised hate groups, or mass-casualty attacks.
Content that promotes self-harm, suicide, or eating disorders. We show safe-message resources instead.
Detailed instructions for making weapons, explosives, or hard drugs.

3.3 Hate & harassment

Slurs and dehumanising content targeting people based on race, ethnicity, national origin, caste, religion, disability, gender, gender identity, sexual orientation, or serious disease.
Bullying, dogpiling, doxing, threats, sustained insults, sexual harassment.
Sharing someone's private information without consent (address, phone number, ID, location).

3.4 Sexual content

No pornography or explicit sexual content in public sanctuaries, on profiles, or in clips/stories.
No non-consensual intimate imagery (NCII / “revenge porn”); reported and removed immediately.
Adult sanctuaries (where the feature is enabled in your region) may allow nudity in artistic / wellbeing contexts only, never sexual solicitation, never minors, never deepfakes.

3.5 Deception

No scams, phishing, fake giveaways, pyramid schemes, or financial fraud.
No deceptive AI-generated content (deepfakes of real people, fake events, manipulated audio passed off as real). AI-generated content involving real public events or people must be clearly labelled.
No election interference or coordinated mis/disinformation campaigns.

3.6 Regulated, illegal & spam

No sale of regulated goods (firearms, drugs, prescription medication, endangered species, stolen goods).
No malware, hacked data, credentials, or services to bypass platform security.
No spam, repetitive content, or bulk messaging.

4. Voice rooms

The same content rules apply to live voice as to posts.
Hosts are responsible for moderating their rooms; we provide mute, kick, and report tools.
Recording happens only when the host explicitly enables it; participants are clearly notified before any recording starts.
Voice rooms today are relayed through our servers (LiveKit SFU) and are encrypted in transit but not end-to-end. We are working on end-to-end encryption for voice and video calls; see Privacy Policy §9.
Persistent abuse leads to room closure and host strikes.

5. Direct messages

DMs are private — we do not read them proactively, we do not use them to train advertising or recommendation models, and we do not sell their content.
DMs are stored on Bloom servers and are not end-to-end encrypted. Reported messages can be reviewed by trained T&S reviewers under strict access controls. See Privacy Policy §9 for the full encryption picture.
Reported messages are reviewed by trained reviewers under strict access controls.
Unsolicited sexual content, harassment, or threats in DMs are violations and may result in immediate account action.
You can block, mute, or restrict any user at any time.

6. How moderation works

Detection — automated systems flag potential violations; users can also report content.
Review — trained human reviewers assess context. Automated systems alone do not make permanent decisions on accounts.
Action — outcomes range from a warning, content removal, feature limits, temporary suspension, to permanent ban. Severity scales with intent, harm, and history.
Notification — you are told what happened, why, and how to appeal.
Appeal — submit an appeal in-app within 30 days. Appeals are reviewed by a different reviewer.

7. Strikes & consequences

Tier 1 (e.g. minor incivility) — warning, content removed.
Tier 2 (e.g. harassment, spam) — temporary feature limits or 7-day suspension.
Tier 3 (e.g. hate speech, repeated harassment, NCII) — long suspension or permanent ban.
Tier 4 (e.g. CSAM, credible threats of violence, doxing) — immediate permanent ban, device-level ban, and report to law enforcement where required.

8. Reporting

Every post, profile, persona, sanctuary, voice room, and message has a Report action. Reports are anonymous to the reported user. Urgent safety issues (imminent harm, child safety) can also be emailed to safety@bloomspace.app.

9. Special protections for minors

Bloom requires every account to confirm a date of birth at sign-up and hard-blocks under-13s. We do not operate a verifiable parental-consent flow, and we do not knowingly collect personal information from children under 13. Accounts identified as belonging to users aged 13–17 are subject to the following non-overridable safeguards:

Inbound message-requests from users who do not share an active sanctuary with them are blocked at the server.
Default messagePrivacy is locked to “mutual sanctuary” and cannot be relaxed.
In-app purchases and subscriptions are not available on the account.
No targeted advertising. Ever.
Reduced notification frequency and a daily-time-spent reminder.
Additional content filters that cannot be turned off.

10. Advertising & commercial activity

Bloom does not run third-party advertising and does not allow user-driven paid promotions in public sanctuaries without our written approval.
Creators may share links to their own work; affiliate spam, dropshipping links, and crypto pump schemes are prohibited.

11. Accessibility

We aim to meet WCAG 2.2 AA on the website and to follow Apple's and Google's platform accessibility guidelines in the app (VoiceOver, TalkBack, Dynamic Type, reduced motion). Please report accessibility issues to access@bloomspace.app.

12. Compliance with platform guidelines

Bloom is designed to comply with the Apple App Store Review Guidelines (including 1.1 Objectionable Content, 1.2 User Generated Content, 1.4 Physical Harm, 5.1 Privacy) and the Google Play Developer Program Policies (including User Generated Content, Inappropriate Content, Child Safety Standards, and Families policies). Where this policy and platform rules differ, the stricter rule applies.

13. Government requests & transparency

We respond to lawful, properly-served government requests for data and publish a transparency report twice a year showing what we received and how we responded.

14. Changes to this policy

We update these rules as the community grows and as new harms appear. Material changes will be announced in-app at least 14 days before they take effect, except for changes needed for urgent safety reasons.