← Back to teacups

Community Reviewers

Last updated: May 19, 2026

Some of the people who use teacups also help moderate it. They are not Tilted Planet employees. They are your neighbors — opted-in, anonymous, and asked to make small decisions about content they didn't write. This page explains what reviewers do, what they don't do, and how to do it well if you become one.

What a community reviewer is

Why reviewers exist

Our automated systems are good at the obvious — illegal content, spam, raw threats. They are not good at "is this person being a jerk on purpose, or is this just a tense conversation?" That's where reviewers come in. Distributed, anonymous, in-the-zone neighbors making small calls about tone and intent are how we keep teacups feeling like a real place instead of an algorithm-policed one.

The reviewer pledge

The decision framework

When you're shown a reported message, ask three questions in order:

1. Is this illegal?

Child sexual abuse material (CSAM), credible threats, terrorism content, non-consensual intimate imagery, hate speech that calls for harm to a protected group. If yes → vote Remove.

2. Does it target a specific identifiable real person for harm?

Named-and-shamed, harassed by handle, deepfaked, hot-or-not'd. Public-figure criticism is allowed; targeting your neighbor by name to encourage harassment is not. If yes → vote Remove.

Important nuance on personal information: teacups is geographically local. "Frank lives above the pizzeria" inside Frank's own block teacup is conversation, not doxxing — Frank's neighbors already know. What makes the same information harmful is either (a) incitement attached to it ("...go break his windows") or (b) posting it to a teacup with no local context — the World tier, or a teleported teacup far from where Frank lives — where the audience has no reason to have it. Vote on what's attached to the information, and on where it's being shared, not on the mere fact of sharing it.

3. Is it just rude, edgy, or offensive — but not illegal and not targeted?

If yes → vote Let it stand. Even if you wouldn't say it. Even if it's wrong. Even if it's gross. The reporter can mute, block, or scroll. The heat system tracks patterns of repeated nastiness across many reports — your job isn't to police tone in this one message.

Examples

These are illustrative. Real reports vary; trust the framework, not the example.

"This new traffic camera is bullshit. The city council can eat my ass."

Political opinion + profanity + crude language. Not illegal, not targeting an identifiable person, no threats.

Vote: Let it stand allow

"Karen at 412 Maple Ave is the worst. Here's her number: 555-0123. Someone make her life hell."

The violation isn't the address — Karen's neighbors plausibly know where she lives. The violation is "someone make her life hell" — incitement attached to identifying information.

Vote: Remove remove

User A: "Best pizza around here?"   User B: "Frank's, 312-111-1111, 123 Fake St"

A local recommendation in a local teacup. Pattern looks like personal info, but it's a business in the neighborhood being recommended to neighbors. Not a violation.

Vote: Let it stand allow

Posted in the World tier: "My ex Sarah lives at 88 Birch Lane in Tulsa, here's her number"

Personal info about a specific person, posted to an audience that has zero local context. The protective effect of "her neighbors already know" doesn't apply when the audience is the whole planet.

Vote: Remove remove

"I had a really bad miscarriage last week and I'm not okay."

Sensitive personal disclosure. Not a violation of anything.

Vote: Let it stand allow

"All [ethnic slur] should be rounded up."

Hate speech calling for harm against a protected group.

Vote: Remove remove

"Your mayor is an idiot and corrupt."

Criticism of a public figure. Strong but legal.

Vote: Let it stand allow

Same user, fifth time this week, posting things like "this block is full of losers"

Individually mild. Pattern is hostile. The heat system handles patterns automatically — your single-message vote should reflect this single message, not the pattern.

Vote: Let it stand heat handles patterns

A photo that's clearly of a minor in a sexual context

CSAM. This should have been blocked by our content filter. If it reached you, vote and report immediately.

Vote: Remove remove + escalates to NCMEC

Reviewer safety

If something nasty reaches you, we're sorry

Our intent is to use technology from Hive.ai and the NCMEC to keep the worst of what gets reported from ever reaching you. Child sexual abuse material, severe violence, sexual abuse, and similar content is supposed to be caught and removed by these automated systems before any reviewer is asked to look at it. We invest in those filters specifically so that the work you do is bounded — borderline tone calls, not trauma.

If something awful gets through, we're sorry. Tell us via support@teacups.com when this happens — every miss helps us tighten the next layer. And please take a break, talk to someone, step away. You don't owe us your wellbeing.

Rewards

How to become one

Reviewer invitations are earned, not requested. Once you've participated in teacups for a while and earned enough tea leaves through normal use, the app will offer you an invitation. There's no application, no interview, no background check — being an active neighbor is the qualification. Accept the invitation if you want to help; decline if you don't. If you decline, you can choose Not Right Now (we might invite you again later) or Don't Invite Me Again (we won't). If you opt out and change your mind, a toggle appears in Preferences to re-enable invitations.

If you're already a reviewer, you can step away anytime — Resign as Reviewer is in both the chat menu and Preferences. After a short break you'll see Rejoin Community Reviewers in the chat menu.

Contact

Questions about your reviewer role: support@teacups.com