There’s a particular kind of whiplash in watching LinkedIn spend three years handing everyone a free AI co-writer, then turn around and build a button so you can dob in the exact content it told you to make. The platform was pretty quick to jump on the AI bandwagon and get it integrated into how posts and comments were crafted. But then it too soon almost ruined the LinkedIn experience.
Now, LinkedIn is scrambling to unscramble the AI mess it created.
Here’s how we got here.
Back in March 2023, LinkedIn expanded its generative AI assistant into profile writing and recruitment ads. By July that year, it had rolled the same idea into the post composer itself: a “help me write” prompt sitting right there in the box, ready to turn a half-formed thought into a fully-formed post.
The pitch was democratisation: nobody has an excuse not to post anymore. Writer’s block? Solved. Blank page? Solved. What LinkedIn didn’t mention is the more obvious incentive beneath that: more posts mean more feed inventory, which means more scrolling, which means more time on the platform. Every AI-assisted post was also a small win for LinkedIn’s own engagement numbers.
The AI slop of it all
So people posted. And a lot. And because the friction to post had dropped to basically zero, what showed up in the feed started looking suspiciously the same: “I’m humbled to announce” openers, the same three-line-paragraph-for-dramatic-effect format, the same hollow “thought leadership” that reads like it was generated because, well, it was.
Then it stopped being just the posts. The comments followed: AI-written replies, comment pods, straight-up automation built to farm engagement rather than say anything of value. LinkedIn’s own numbers from mid-2026 give you a sense of scale: the platform says it’s now blocking hundreds of thousands of automated comment attempts a day, and millions more over recent months. And LinkedIn isn’t even a special case here – Cloudflare has reported that bot traffic now outstrips human traffic across the web generally. LinkedIn’s slop problem is a symptom of a much bigger one.
If you’ve spent any time in your own feed lately, you don’t need the stats explained to you. You already know what we’re talking about. You’ve already scrolled through it. And it’s been harrowing, right?
LinkedIn turns on its own creation
At some point, and without a big splashy announcement, the algorithm started pushing back. Content that read as obviously AI-generated began losing reach, while posts with a distinct, original voice made it up to the top of the feed. LinkedIn never framed it as “sorry, we made a mistake handing out the AI pen.”
But that’s functionally what the change amounts to: the platform penalising the exact behaviour its own tool created.
The irony is a hard one to miss.
Then they cracked down: the "seems like AI slop" button
On 30 July 2026, LinkedIn made it official. It launched a “seems like AI slop” button, letting users flag posts they suspect are AI-generated. This felt like the most wild move a social media platform has made in recent years that I initially thought it was made up.
That feedback trains LinkedIn’s detection models, which then suppress similar content showing up in recommendations from outside your network.
In the same announcement, LinkedIn killed off “enhance your post” – the feature that would rewrite your post for you – and replaced it with a proofreading tool that corrects grammar and clarity without touching your voice. LinkedIn didn’t decide to walk away from AI altogether. It’s just very specifically walking back the version of AI that replaces people’s thoughts with that generic AI stuff.
LinkedIn’s Chief Product Officer put it plainly: people come to LinkedIn “to connect with real people and share their real perspectives.” Which, fairly said, is exactly what the platform stopped optimising for the moment it started auto-generating everyone’s perspective for them.
The takeaway? Stop chasing the trend, stick to what's actually good
It’s tempting to read all this as a story about AI specifically. We don’t think that’s the real lesson. The pattern here is one we’ve seen before and will most probably see again: a shortcut shows up, everyone piles onto it at once because it’s easy and everyone else is doing it, the platform gets flooded with identical low-effort content, and then the algorithm… or the audience… or both turns on it.
The accounts that never had a reach problem through any of this are the ones that weren’t chasing the shortcut in the first place. They were posting things with an actual point of view, in their own words, at a pace they could sustain without a machine doing the thinking for them.
The practical takeaway isn’t “avoid AI tools forever.” It’s: don’t build your content strategy around whatever trick is currently working.
Quality over quantity, every time. One genuinely useful, specifically-you post beats five generic ones optimised for a loophole that’ll be closed or all clogged up in six months. Know what good content looks like for your brand and stick to it, regardless of what everyone else’s feed is doing this quarter. LinkedIn’s AI slop war is a case study in what happens to the businesses that don’t.
—
Need a hand with your social media? We’d love to help with that 👋
