Quick answer: Twitch does not publish a complete list of “banned words.” A word can be allowed in one context and violate policy in another. Hateful slurs, targeted harassment, sexual harassment, credible threats, doxxing, scams, and other prohibited conduct can lead to enforcement, while each channel may set stricter chat rules of its own.
This guide was substantially rewritten in July 2026 from Twitch's current Community Guidelines and moderation documentation. It does not reproduce slurs or promise that any particular word is “safe.”
There is no official Twitch banned-words list
Twitch says it does not make its prohibited-slur list public because publishing it could help people evade enforcement. Twitch also evaluates context. That means a copied list from a blog, forum, or chatbot cannot tell you with certainty what Twitch will allow.
The practical rule is to evaluate the conduct, target, intent, and context—not only the vocabulary. Twitch's current Community Guidelines are the primary source when a policy question matters.
Speech and conduct that can violate Twitch policy
Hateful conduct
Twitch prohibits conduct motivated by hatred, prejudice, or intolerance toward protected characteristics. Its policy covers discrimination, denigration, harassment, and violence. Replacing letters, using an emote, or implying a slur does not make hateful conduct acceptable.
Harassment and sexual harassment
Targeted personal attacks, encouraging harm, malicious brigading, repeated unwanted behavior, sexual requests, sexual objectification, and other unwanted sexual conduct can violate policy. Context such as whether behavior is targeted, repeated, threatening, or clearly unwanted affects enforcement.
Threats, self-harm encouragement, and doxxing
Credible threats and encouragement to harm oneself or others are not ordinary “trash talk.” Do not reveal private addresses, precise locations, government identifiers, sensitive financial details, or other personally identifying information with malicious intent.
Sexually explicit content and prohibited promotion
Twitch prohibits pornographic content and certain explicit sexual material, including promoting or directly linking to it in broadcasts, profiles, or chat. Content labels do not turn prohibited content into permitted content; they are used for content Twitch allows but requires to be labeled.
Impersonation, fraud, and malicious links
Scams, phishing, impersonation, malware, and attempts to manipulate Twitch systems are policy and security issues even when the chat message contains no offensive word. Moderators should treat an unknown link as a behavior and safety question, not merely a vocabulary check.
Profanity is not the same as hateful conduct
Twitch's AutoMod treats profanity as a moderation category, but that does not mean every swear word automatically causes a platform suspension. A channel can choose a family-friendly standard stricter than Twitch's baseline. At the same time, ordinary-looking words can become harassment when they are used to target or intimidate someone.
Do not use “it was only a joke” as a safety test. Twitch permits genuine educational or satirical discussion in some circumstances, but says that disguising hateful viewpoints as comedy or education can still lead to enforcement.
Twitch policy and channel rules are different layers
- Twitch rules: apply across the service and can result in account-level enforcement.
- Channel rules: are set by the broadcaster and can be stricter, such as no profanity, no spoilers, or no political discussion.
- Moderator decisions: apply those rules in a specific chat and may include warnings, deleted messages, timeouts, or channel bans.
A phrase can therefore avoid a Twitch suspension but still earn a timeout or ban from a particular channel. Read the channel's rules before participating.
How to configure AutoMod
- Open Creator Dashboard → Moderation → Settings.
- Turn on AutoMod and select a starting level.
- Adjust discrimination and slurs, sexual content, harassment, profanity, and Smart Detection separately if needed.
- Add channel-specific blocked terms and phrases.
- Add permitted terms only after reviewing false positives.
- Tell moderators what standard you expect and when to warn, timeout, ban, or report.
According to Twitch's current AutoMod guide, AutoMod holds a flagged message for review; it does not itself timeout, ban, or mute the user. Moderators decide whether to allow or deny the message and can take a separate action.
Use blocked terms without overblocking chat
- Block phrases that expose personal information, repeat known abuse, or match recurring spam.
- Prefer a multi-word phrase when blocking a common word would catch harmless discussion.
- Keep sensitive blocked terms private when the list itself would reveal information.
- Review permitted terms carefully; an exception should not bypass another harmful part of a message.
- Revisit the list after real incidents instead of downloading a giant generic blocklist.
Twitch documents public and private blocked terms. Private terms are visible to the channel owner rather than the moderator team, which can be useful for protecting personal information.
What to do when a viewer crosses the line
- Preserve context in the native viewer card or Mod View.
- Delete the message or deny it in AutoMod if it should not appear.
- Use a warning or timeout for behavior that may be correctable.
- Ban repeat or severe offenders from the channel.
- Report conduct to Twitch when it may violate platform policy; report the specific message when possible.
- For an immediate, credible safety threat, contact the appropriate local emergency or law-enforcement service rather than relying only on platform moderation.
Twitch says broadcasters are expected to make good-faith efforts to mitigate hateful or harassing material that viewers put on stream. AutoMod, timeouts, bans, and reports help demonstrate that the channel is actively moderating rather than encouraging abuse.
Frequently asked questions
What words get you banned on Twitch?
There is no reliable universal word list. Hateful slurs and other language used as hateful conduct, harassment, threats, or prohibited sexual content can result in enforcement. Twitch assesses context and does not publish its complete prohibited-slur list.
Can you swear on Twitch?
Profanity is not automatically the same as hateful conduct, but channel rules can prohibit it and context can turn language into harassment. Content labels and audience expectations may also apply.
Can you say a slur while quoting someone?
Do not assume quotation makes it safe. Educational context may matter, but Twitch still evaluates how the material is presented and whether it promotes or targets abuse. Paraphrase when the exact term is unnecessary.
Can streamers be punished for what appears in chat?
Twitch expects streamers to mitigate prohibited third-party content. Its guidelines say good-faith moderation matters, while failing to use available tools against hateful conduct can lead to enforcement.
Where should I check for policy changes?
Use Twitch's Community Guidelines, Safety Center, and Help Portal. A dated third-party “banned words list” should never be the final authority.
