Blog

Platform News Platform News

LinkedIn Is Down-Ranking AI-Generated Comments: What Brands and Executives Should Do Now

LinkedIn has started pushing AI-generated comments down the page. Here's what changed in November and how I'd rebuild a brand's LinkedIn engagement plan around it.

LinkedIn is now actively pushing AI-generated comments down the page. In a November update, the company said it has retrained the systems that rank comments so that replies its classifiers judge to be machine-written and low-relevance appear less often under the default "most relevant" view. For brands and executives who have been using automation to "stay active" on LinkedIn, this is the moment to stop.

Here is what changed, and how I'd adjust a brand's LinkedIn engagement plan because of it.

What LinkedIn announced

LinkedIn's Chief Product Officer, Hari Srinivasan, outlined the changes in mid-November:

  • New comment classifiers. LinkedIn updated its detection tools to assess how relevant a comment is to the post and to flag suspected AI replies as less valuable.
  • Re-ranking, not deletion. Suspected AI comments are not removed. They are ranked lower, so fewer people see them.
  • Smarter notifications. Creators should be notified mainly about comments they are likely to care about, which means generic replies are less likely to trigger a notification at all.
  • Early results. LinkedIn said that in testing, views of what it calls "slop" comments fell by more than 20%. The update is set to reach all users soon; no exact date was given.

This follows LinkedIn's August move to let members report posts that "seem like AI slop," which LinkedIn said more than a million people used in the first three weeks. We covered the post side of that crackdown in LinkedIn's AI slop crackdown: what brands should do now. Comments were the obvious next step.

Why this matters more than it sounds

Comments are where much of LinkedIn's real business value happens. A thoughtful comment from a founder or a head of product on an industry post is often seen by more of the right people than that person's own posts. It is how executives get discovered by peers, buyers and journalists, and it is how a brand's people show up in conversations they don't own.

That is exactly why comment automation spread. Tools that generate "Great insight, thanks for sharing!" or a three-sentence summary of the post made it easy to look active on dozens of posts a day. LinkedIn is now telling everyone that this activity is being quietly discounted. If your team or your executives rely on it, a lot of that effort is now buying very little visibility, and it risks making your people look like bots to the exact audience you want to impress.

What I'd change in a brand's LinkedIn plan

1. Audit who is automating

Start by finding out. Ask your social team, your executives' assistants and any ghostwriting or growth agencies whether comments are being written or posted by a tool. It is not unusual for an executive to be unaware that their account has been leaving automated comments every day. Turn it off, or at minimum turn it into a drafting aid that a human rewrites.

2. Fewer comments, much better ones

Replace volume targets ("comment on 20 posts a day") with quality targets. A comment worth making adds something the post didn't have:

  • a specific example from your own experience,
  • a respectful disagreement or a caveat,
  • a data point you can actually stand behind,
  • or a question that moves the conversation forward.

Five comments like that a week will do more for an executive's reputation than a hundred generic ones.

3. Build a short "who we engage with" list

Engagement works best when it is targeted. For each executive who is active on LinkedIn, list the 20–30 people whose posts matter most: key customers, partners, analysts, journalists, respected peers. Check their activity a few times a week and comment when you have something real to add. This is the human version of what automation was pretending to do.

4. Respond well on your own posts

With notifications filtering out low-value replies, the comments you do receive are more likely to be worth answering. Reply quickly and properly, especially in the first hours after posting. Conversations in your own comment section are a strong signal that the post is worth showing to more people.

5. Use AI where it helps, not where it shows

I am not against AI in social workflows. It is useful for research, summarising a long report before you comment on it, checking tone, or proofreading (which is now how LinkedIn itself describes its post tool). The line is simple: if the words people read under your name were not written or genuinely rewritten by you, they are a liability.

The bigger pattern

The direction is clear. LinkedIn is reducing the reach of generic AI posts and comments while placing more weight on recognisable people with real expertise, which ties into how LinkedIn content is now showing up in AI search answers. We wrote about that in LinkedIn and AI search visibility. For B2B brands, the takeaway is that your executives' authentic voices are an asset worth protecting, and the shortcuts are being closed one by one.

If you're rebuilding your LinkedIn approach from the ground up, our guide to LinkedIn strategy for B2B companies walks through the full plan, from company page to executive presence. And if you need someone to set the rules and coach your leaders through it, that is the core of our social media strategy consulting and team coaching.

Sources

Social Media Today, "LinkedIn cracks down on AI-generated comments" (November 15, 2026); Dataconomy, "LinkedIn cuts visibility of AI-generated comments" (November 16, 2026). Details on the August "seems like AI slop" reporting option are from the same coverage.


Want a second opinion on your social?

I offer a free social audit for brands: send me your handles, I review what you're doing, and we talk through what to keep, cut and try next. Request your free social audit →

Read More