How to Block Bots and Spam on X (Twitter) in 2026

How to spot a bot in 2 seconds

Most bots have the same fingerprints. Once you know them, you spot them instantly:

• Generic username with random numbers — @user12894, @cryptoking2847.
• Default avatar or AI-generated face — slightly waxy skin, perfect symmetry, blurry background.
• Bio with crypto address or Telegram link — almost always a scam.
• Reply doesn't match your post — a bot replying "great thread!" to a 4-word tweet.
• Account created in the last 30 days — check the join date on their profile.
• Massive follow count, tiny following — or vice-versa. Real users are roughly balanced.

Two of these together = bot. Three or more = block without thinking.

X's built-in Safety Mode

X has a feature called Safety Mode that auto-mutes accounts using "potentially harmful language." It's free, opt-in, and easy to set up:

1. Open Settings and privacy
2. Tap Privacy and safety
3. Tap Safety
4. Tap Safety Mode
5. Toggle on, set duration (default 7 days, repeatable)

Safety Mode is good for one specific thing: temporarily muting waves of harassment after a viral post. It's not a general bot filter — it can't catch crypto spam or polite-sounding scam replies. Use it as one layer, not the only one.

Block vs Mute vs Report — which to use

All three are different tools. Use the right one:

• Mute — silent, low-friction. They don't know. Their posts and replies disappear from your feed. Good for: random noise, accounts that are just annoying.
• Block — they can't see your profile, can't reply to you, can't follow you. They do know they're blocked if they check. Good for: harassment, persistent spammers.
• Report — affects X's actions. Reports for spam, fake account, or terms violations get reviewed. Good for: actual scam bots, impersonation, illegal content.

Default to mute for noise, block for real problems, report only for genuine violations. Reporting trivial things waters down the system.

Bulk-blocking with shared lists

There are tools that let you import shared block lists — for example, lists of known crypto bots maintained by the community. Risk: false positives. Imported lists can include real users misidentified as bots.

If you want to try this approach: use an open-source, actively maintained list (search GitHub for "X bot block list"). Read the criteria. Accept that you may block 5–10 real people and have to manually unblock them later.

For most users this is overkill. Manual blocking + a good extension is usually enough.

The silver bullet: engagement thresholds

Spam bots almost always have very low engagement. Their replies get 0–3 likes because no one finds them useful. By hiding tweets and replies below a minimum engagement, you remove most bot output without listing a single keyword.

This isn't a feature X offers natively. You need a browser extension. The setup is simple:

1. Install an extension that supports engagement filtering (X Filter Pro, ControlPanel, others)
2. Set a threshold — start with "hide replies below 5 likes"
3. Watch the noise drop

On a busy thread that used to have 200 replies, you might end up seeing 20 — and they'll all be from real people whose replies actually got engagement. Bots are simply gone.

Browser extensions that help

X Filter Pro — engagement-threshold filtering, reply-spam removal, and AI-based bait detection. Free tier covers basic filtering; Pro tier ($2/mo) adds engagement thresholds and unlimited AI.

uBlock Origin — generic ad blocker. Won't catch bot replies but stops bot-related tracking and some sponsored bot content.

ControlPanel for Twitter — toggle UI elements, hide ads, has some bot-related toggles. Free.

Note: avoid extensions that promise to "detect bots with AI" without explaining how. If they read every page you visit, they're a privacy risk. Stick to extensions that limit themselves to x.com.

Don't forget: don't engage

The most underrated bot-defense is also the simplest: don't reply, don't quote-tweet, don't even react. Engagement is what trains the algorithm to show you more bots. A muted account with a strong filter and zero engagement on bot replies will see less spam over weeks.

Quick checklist

1. Turn on Safety Mode (Settings → Privacy → Safety)
2. Block obvious crypto bots manually — spend 5 minutes once a week
3. Mute generic noise instead of blocking; reserve block for real problems
4. Report actual scam/impersonation accounts (helps platform-wide)
5. Install an extension with engagement-threshold filtering
6. Set engagement threshold to 5 likes for replies; adjust over time
7. Don't reply to bots — engagement trains the algorithm to show more

That's the playbook. X's native tools handle some of the work; engagement-based filtering handles most of the rest. The combination removes 90%+ of bot exposure with minimal effort once set up.