How to hide AI posts on LinkedIn (2026 guide)

LinkedIn does not have a setting that hides AI generated posts. It gives you three tools: Not interested on a single post, Hide posts from an author, and Unfollow. They work on single posts and on single people. None of them looks at what a post actually says. To filter the slop itself, you need a tool that reads each post as you scroll and checks it against rules you write.

This guide covers what LinkedIn's own controls do and when they are enough. It then explains how to filter by rule for everything they miss.

Why your LinkedIn feed fills up with AI posts

A lot of what you scroll past on LinkedIn was at least drafted by a model. Originality.ai, a company that sells an AI detector, ran its detector on longer English posts. It estimated that more than half were likely AI generated. Treat that number with care, because the company that published it also sells the detector. Still, the trend it shows matches what most people see in their own feeds.

AI is not really the problem, though. A thoughtful post someone wrote with help is fine. What wears you down is the format around it:

• The one line hook ("I got rejected from 47 jobs. Here is what I learned.")
• Short paragraphs stacked like a poem
• A lesson anyone could have guessed
• A closing question that exists to farm comments

Those posts get reactions, reactions buy reach, and so the feed rewards more of them. Merriam-Webster even made slop its 2025 word of the year, defining it as low quality digital content produced in quantity by AI.

What LinkedIn's built in controls can do

LinkedIn documents its feed controls in its help article on hiding content in your feed. Open the menu (the three dots) at the top right of any post to find them.

Not interested

Not interested removes the post and tells LinkedIn you want fewer posts like it. Use it on the ones that are pure template: the reposted listicle, the "comment YES for the guide" post. It is a hint to the ranking system, not a rule. It works best when you use it often and consistently.

Hide posts from a person or company

Hide posts from [name] works on the author. It keeps your connection and stops their posts from showing in your feed. Use it for the colleague who posts three times a day and who you still want to message.

Unfollow

Unfollow also keeps the connection but drops the person or Page from your feed. They are not notified. You can review everyone you have unfollowed later, under the feed preferences in your settings.

Stop seeing other people's activity

A lot of slop reaches you secondhand: "Jordan commented on this", "Priya likes this". When the same connection keeps pulling viral posts into your feed, unfollowing them usually clears out their reactions along with their own posts.

Where the built in controls stop working

These tools work on authors, not on content. That leaves three gaps.

1. New authors keep arriving. LinkedIn shows you posts from people you do not follow, based on what your network reacts to. Hiding one template poster does nothing about the next one with the same template.
2. Good people post slop sometimes. You may want a colleague's job updates but not their weekly motivational carousel. Unfollowing loses both.
3. Nothing reads the text. LinkedIn does not document any keyword or topic filter for the feed. You cannot tell it "hide posts that end with Agree?" or "hide humblebrags about waking up at 4am".

The result is a cleanup you have to do again every week. After a few days of not clicking Not interested, the feed drifts back.

How to filter LinkedIn posts by what they say

The only way to filter on content is a browser extension that reads posts on the page as you scroll. Two kinds exist.

Keyword blockers

Keyword blockers hide any post that contains a word or phrase you list. They are simple and they run entirely in your browser. They are also blunt. Block "AI" and you lose the useful post about your industry's AI policy. Block "Agree?" and the bait comes back next week as "Thoughts?". Template slop is a pattern, not a vocabulary, so a word list misses most of it.

Rule based filters that read the whole post

The other approach judges each post against a description of what you consider slop, written in plain language. It catches the pattern rather than the words. "Asks for comments without giving anything in return" covers "Agree?", "Thoughts?", "Comment GUIDE below", and next month's variant too.

That is how SlopStamp works. You describe slop in your own words, or start from a default policy. The default covers engagement bait, recycled templates, generic motivation, ragebait, unsupported expertise, and disguised ads. As you scroll LinkedIn, X, or Reddit, each post is checked against your rules. Posts that match are hidden or stamped with the reason. Nothing is deleted from LinkedIn, and every hidden post is one click away if you want to read it after all.

A setup that keeps your feed clean

Combine the two layers so each does what it is good at.

1. Unfollow the obvious accounts once. Five minutes with LinkedIn's own tools removes the people you never want to hear from.
2. Use Not interested for a week. It nudges the ranking system and costs nothing.
3. Write rules for the patterns. Put the content you are tired of into words: engagement bait, humblebrags, AI listicles, recruiter spam. Let a filter apply those rules to every post, from every author, including people you have never seen before.
4. Keep exceptions for what you value. If you always want hiring posts or posts from your own company, say so, so the filter leaves them alone.
5. Review what was hidden now and then. A good filter shows you what it caught and why, so you can loosen a rule that is too strict.

Test a post before you change anything

Not sure whether a post you keep seeing counts as slop? Paste it into Slop Check. It is free and needs no account. You get a verdict and a one line reason, using the same default policy SlopStamp starts with. That is a quick way to find out whether a filter would agree with you before you install one.

If the posts that annoy you are mostly comment bait, read LinkedIn engagement bait: 8 patterns to spot. For a clear definition of the term itself, see What is AI slop?.