LinkedIn is taking its AI content problem seriously enough to give users a dedicated button for it. The professional platform has begun testing a new reporting option allowing members to flag content that "seems like AI slop." Simultaneously, LinkedIn is pulling back on the generic writing tools that helped accelerate the problem in the first place.
TL;DR
- LinkedIn is testing a "seems like AI slop" report option that privately flags a post to its author via analytics — no public label.
- Flagged posts get their algorithmic reach throttled, similar to a "not interested" click.
- LinkedIn is removing its "rewrite with AI" button and replacing it with a lighter proofreading tool that fixes grammar without changing your voice.
- A Pangram study found over 40% of long-form LinkedIn posts are fully AI-generated — the most of any major platform.
- LinkedIn also says it has blocked billions of automated posting attempts and caught hundreds of thousands of automated comments in the last few months.
- Net effect: sounding generic is now a liability on LinkedIn, which favors tools like Postessia that write in your actual voice instead of generic AI templates.
What the new tool actually does
When someone flags a post this way, the author doesn't get a public callout. Instead, it shows up privately in their analytics dashboard as a signal that some readers felt the post came off as inauthentic or overly AI-assisted. LinkedIn's Chief Product Officer, Hari Srinivasan, described it as a private nudge rather than a public shaming mechanism.
But it's not just cosmetic. A flagged post's algorithmic reach gets throttled in roughly the same way a "not interested" click would — quietly cutting its visibility, not deleting it or labeling it publicly for other viewers.
LinkedIn is also stepping back from its own AI-writing push
Here's the more interesting part: LinkedIn is removing the "rewrite with AI" button it had promoted across the platform. That feature, which let people run their drafts through an LLM to "enhance" them, is being replaced with something more conservative — a proofreading assist that fixes grammar and clarity without rewriting a person's voice into generic LinkedIn-speak.
It's a quiet admission that the platform's own tools were part of why so many posts started sounding identical.
The scale of the problem
This isn't a minor tweak. A recent study by AI-detection company Pangram found LinkedIn to be the most AI-saturated major social platform, with over 40% of long-form posts flagged as fully AI-generated. That's a huge share of the feed, and it explains why LinkedIn has been layering fix after fix on top of each other — an earlier update this year already worked to reduce the reach of AI-generated content, and this reporting tool builds directly on that effort.
The company says its detection systems have also improved enough that people should simply start seeing fewer AI-flavored posts recommended to them, independent of user reports. And the bot problem is apparently massive on the back end too — Srinivasan said LinkedIn caught hundreds of thousands of automated comments and blocked billions of automated posting attempts in just the last few months.
Will it actually change anything?
That's the open question. A reporting button and a softer writing assistant are useful signals, but they're working against a flood — millions of posts a day, a huge chunk of them AI-written or AI-bot-generated, and a format (the "I failed, then I learned" post) that was already formulaic long before AI got involved.
What this update really does is make the platform's stance explicit: sounding like everyone else is now a liability, not a safe default. For anyone posting on LinkedIn, the incentive just shifted further toward writing that actually sounds like a specific person, not a template running on autopilot.
That distinction is exactly what's driving the next wave of AI writing tools too. Generic "rewrite this for LinkedIn" prompts are precisely what's now getting flagged — so the tools that survive this shift will be the ones built to sound like the person using them, not like every other AI-assisted post in the feed. That's the problem Postessia was built around: instead of generating generic AI copy, it clones a founder's actual voice and turns their raw thoughts into posts that read like they wrote it themselves. As LinkedIn tightens the screws on anything that reads as templated, that difference — voice versus template — stops being a nice-to-have and starts being the whole game.