Moderation and safety
What is content moderation?
Content moderation is reviewing and acting on user posts to keep a community safe and on-topic. Learn how it works on Discord and why it matters.
Content moderation is the practice of reviewing what people post in an online space and deciding whether it stays, gets removed, or leads to action against the poster. On Discord, it covers messages, images, links, usernames, and voice behavior. The goal is to keep a community safe and on-topic according to rules the community has written down. See also the general overview on Wikipedia.
Why does content moderation matter?
An unmoderated community tends to drift toward its worst behavior. Spam buries real conversation, scammers target newcomers, and a few hostile members make everyone else leave quietly. Moderation protects the people who would otherwise just disappear, and it gives members a predictable experience: they know what is allowed and what happens if they break the rules.
What are the main types of content moderation?
Moderation systems differ mainly in when a decision happens and who makes it.
| Type | When it happens | Who decides | Typical use |
|---|---|---|---|
| Pre-moderation | Before content is visible | Human or software | Verification gates, approval queues |
| Post-moderation | After content is posted | Human or software | Most Discord servers |
| Reactive moderation | After a member reports it | Humans | Reports and flags |
| Automated moderation | Immediately or in batches | Software rules or AI | Spam, links, clear abuse |
| Hybrid | Software flags, human decides | Both | Larger servers |
Most healthy servers use a hybrid: automation handles the obvious and repetitive cases, and humans handle context, appeals, and judgment calls.
What does a moderator actually do on Discord?
Day to day, moderation is a small set of repeatable actions, roughly in order of severity:
- Delete a message that breaks a rule.
- Give a warning, privately if possible, with the rule it broke.
- Time out a member so they cannot post for a set period.
- Kick a member, who can rejoin with a new invite.
- Ban a member, who cannot return unless unbanned.
Good moderators match the response to the behavior. A first-time off-topic post deserves a gentle reminder, not a ban. Repeated harassment or a scam link deserves something much firmer.
What can be moderated?
- Text messages: insults, harassment, hate speech, spam, and off-topic posting.
- Links: scam and phishing links, unwanted server invites, and malware. See how to stop spam and scam links on Discord.
- Images and files: explicit or abusive images.
- Behavior over time: raiding, ban evasion, and repeated low-level disruption.
- New members: gating who gets full access. See how member verification works.
What are the limits of content moderation?
No system catches everything, and no system is perfectly fair. Rules are written in advance but behavior is creative, so there will always be gray areas such as sarcasm, in-jokes, quotes, and reclaimed language. Automated tools in particular trade off two errors: missing real abuse (false negatives) and flagging harmless posts (false positives). Tightening one usually worsens the other. That is why how AI moderation works matters, and why a human should be able to review and reverse automated decisions.
What makes moderation fair?
Fair moderation is consistent, explained, and reviewable. That means written rules, the same rule applied to regulars and newcomers alike, telling people why something was removed, and giving staff a private place to discuss hard cases. Our guide to Discord moderation best practices covers this in detail.
Doing this with Based
Based's automatic moderation module is off by default and defaults to a "Review only" mode that never changes messages or members. Staff can raise the maximum action step by step, and a private log channel records what it did. It never goes past the maximum action you choose. Details are in the automod docs and the AI automod feature page.
Frequently asked questions
What is the difference between moderation and censorship?
Moderation enforces the rules of a specific community that people choose to join. Censorship usually means a controlling authority suppressing speech across a whole society. A server owner removing off-topic or abusive posts is setting house rules.
What is the difference between pre-moderation and post-moderation?
Pre-moderation reviews content before anyone sees it, while post-moderation reviews it after it is published. Most Discord servers use post-moderation because chat is fast, and add pre-moderation only at the door through a verification gate.
Do small Discord servers need moderation?
Yes, but it can be light. Even a small server benefits from a few written rules, one or two trusted moderators, and a way to remove spam quickly.
Can software replace human moderators?
Not safely. Software is good at volume and speed, but humans are better at context and fairness. Most communities use automation to flag or handle obvious cases and keep people in charge of serious decisions.
Keep learning
- Moderation and safety
How does AI moderation work?
AI moderation uses language models to read messages and flag abuse, spam or rule breaks. Learn how it works, where it fails, and how to use it safely.
- Moderation and safety
Discord moderation best practices
Practical Discord moderation best practices: clear rules, tiered responses, least-privilege roles, logging, verification and a staff process that scales.
- Moderation and safety
How to write Discord server rules
Learn how to write clear Discord server rules: what to include, how to word them, consequences, examples, and how to enforce them consistently.
- Moderation and safety
How does member verification work on Discord?
Member verification gates new Discord members behind a check before they can chat. Learn the options, how question-based gates work, and best practices.
Try it in your server
Based is an AI Discord bot in 300+ servers. Every slash command also works with the "." prefix.