How to mute words on X (and where mute does not reach)

#Quick answer

To mute words on X: open Settings and privacy > Privacy and safety > Mute and block > Muted words, tap the plus icon, type the word, then pick where it applies (Home timeline, Notifications, or both), who it applies to, and for how long. In the iOS and Android apps there is a shortcut: the Notifications tab, the gear icon, then Muted words. You can mute words, phrases, usernames, emojis and hashtags, and the mute is invisible to everyone else.

Last updated: October 2026

Muted words are the most underused setting on X (formerly Twitter), and the reason is that most people set one up, still see the thing they muted, and quietly conclude the feature is broken. It usually is not. Mute has a precise scope that X documents and almost nobody reads, and four separate choices attached to every single entry. Below: the exact steps on each platform, what those four choices do, the places a muted word still reaches you, and the one legacy default that silently breaks old mute lists.

#How to mute words on X, step by step

Every platform lands in the same place. Only the route differs.

On x.com in a browser:

1. Click More in the left navigation, then Settings and privacy.
2. Open the Privacy and safety tab, then Mute and block.
3. Click Muted words, then the plus icon.
4. Enter the word, phrase, hashtag, emoji, or @handle.
5. Tick Home timeline, Notifications, or both.
6. Choose From anyone or From people you don't follow.
7. Under Mute timing, pick Forever, 24 hours, 7 days, or 30 days.
8. Click Save.

On iOS and Android, skip the settings menu entirely. Go to your Notifications tab, tap the gear icon, tap Muted words (on iOS, tap Muted first), then Add or the plus icon, and work through the same choices.

Worth knowing before you start: entries go in one at a time. X documents no bulk import and no paste-a-list box, so a 40-word filter is 40 trips through that form. Build the list in order of how much each term annoys you and stop when the feed feels right.

#The four choices on every muted word

This is where the feature is actually configured, and where it goes wrong. Each entry stores four settings, per X's advanced muting options page:

The scope setting is the one that produces the "mute is broken" experience. From people you don't follow exempts your entire following list, which is sensible for notification triage and useless for cleaning a timeline mostly made of accounts you follow. If the goal is a quieter Home timeline, the answer is almost always From anyone.

Duration deserves more credit than it gets. Forever is the default, but the 24-hour and 7-day options are what make mute genuinely practical for a finale, an election, a conference hashtag, or a sports result you want to watch later. Set it, enjoy a clean feed, and let it lapse without maintenance.

#Where muted words still reach you

Mute is a filter on two specific surfaces, not a platform-wide content block. X spells out which surfaces, and the gaps in that list explain most complaints about the feature.

Muting does remove matching posts from your Notifications tab, your push notifications, SMS and email notifications, your Home timeline, and from replies to posts. Recommendations X sends you by email or in-app will not suggest content containing your muted words either.

Here is what it does not do:

• Search is exempt. X's own note is blunt: muting words and hashtags only applies to your notifications and Home timeline, and you will still see those posts via search. Searching a term you muted shows it, which is a feature rather than a bug once you know it. It is also why X advanced search stays usable on topics you have filtered out of your feed.
• Followed accounts can be exempt. Entirely down to the scope setting above.
• Post notifications you opted into override it. If you turned on notifications for when a specific account posts, X documents that muted words will not apply to those. You asked for that account's posts specifically, so the filter stands down.
• It does not reach into DMs or someone else's view. Nothing about muting changes what other people see or what your own posts do.

Know those four and mute stops feeling unreliable. It is doing exactly what the documentation says.

#Why your old muted words stopped working

A specific trap catches long-time X users. Muted words started as a notifications-only feature and later gained the Home timeline option. X applied a default to every entry that predated the change: Notifications only, From anyone, Forever.

So a word you muted in 2019 is still muting your notifications and has never once touched your timeline. Nothing failed. The entry simply never had the timeline box ticked, because that box did not exist when you created it.

The fix takes a minute. Open Muted words, tap each old entry, tick Home timeline, confirm the scope says From anyone, and save. Worth doing in one pass over the whole list rather than one word at a time when you get annoyed.

#The matching rules X actually documents

Four rules govern what a muted entry catches, and two of them save you work:

• Case does not matter. Add CATS and every cats, Cats and cATS is muted too.
• A word covers its hashtag. Mute unicorn and both unicorn and #unicorn are muted. Separate entries for the plain word and the hashtag are redundant, which trims most people's list by a third. Useful context if you are still weighing how to use hashtags on X for your own posts.
• Punctuation can sit inside a phrase, but trailing punctuation is unnecessary. Mute season finale rather than season finale.
• An @handle mutes mentions, not the account. Enter @someaccount and you stop seeing notifications and timeline posts that mention that account, while the account itself stays unmuted. That is the right tool for a name you are tired of hearing about rather than a person you want to stop hearing from.

What X does not document is substring behaviour, so do not count on crypto catching cryptocurrency. When a term matters, add both forms and keep the list honest.

#Mute a word, mute an account, or block

Three tools, three different problems. Picking wrong is why people reach for the heaviest option first.

That last one is separate from muted words and genuinely useful. Open any post in a conversation, tap the three-dot menu, and choose Mute this conversation. New notifications stop, though X notes the posts stay visible in your timeline and when you open the original post. It is the tool for a thread that went sideways under one of your replies, alongside hiding replies on X when the problem is what other people are reading.

Muting an account behaves differently depending on whether you follow it, which trips people up. For a muted account you do follow, replies and mentions still appear in your Notifications tab. For one you do not follow, they stop. Both rules are on X's mute page.

#A mute list worth building

A single mental model makes this setting pay off: muted words are a standing instruction to your notifications, not a reaction to one annoying post. A few categories do nearly all the work.

Spoilers and events. Show titles, team names, match words like final score. Perfect for the 24-hour and 7-day durations.

Engagement bait aimed at your replies. Terms that only ever show up in the posts you regret opening. Set these to Notifications, From people you don't follow, Forever, and your mentions get noticeably calmer without hiding anything your actual audience says.

Your own triggers. Products you will never buy, discourse you have no stake in, the recurring argument of the month. Timeline, From anyone, 30 days, revisit later.

One handle you keep hearing about. The @handle form, Notifications only.

Just as useful is knowing what to leave alone. Avoid broad words that appear in normal conversation. Mute AI and you will quietly delete half your feed, including the replies you wanted. Narrow terms, specific phrases, short durations.

If your notifications are still unusable after all this, the problem may be upstream of mute. X's quality filter and its advanced notification filters work on the sender rather than the words, and how to turn off notifications on X covers those controls plus which ones are on by default.

#Editing, unmuting, and the thing mute cannot do

To change or remove an entry, go back to Muted words, tap the word, adjust Mute from or the mute timing, then save. To remove it entirely, tap Delete word and confirm. Everything is reversible and nothing is announced.

What mute cannot do is make your own posts land better. A clean notifications tab helps you spot the replies worth answering, and replies are where accounts actually grow, but the filter does not write anything for you. If that is the real problem, X-Autopilot drafts replies in your voice from real Chrome on your own Mac, so your session cookie stays on your machine rather than sitting in a cloud service. Be straight about what that is: browser-route automation on X is a gray area rather than a sanctioned path, and running locally on your own residential connection is a smaller detection surface than a cloud tool posting from a shared IP pool, not an absence of risk. X's automation rules are short, and worth reading yourself instead of trusting any vendor's summary.

Honest summary of muted words: a precise tool that behaves exactly as documented, and almost everyone configures it wrong once. Tick Home timeline, set the scope to anyone, use short durations for events, and go fix the legacy entries that have been muting nothing but your notifications for years.