On July 30, LinkedIn did something almost unthinkable for a Microsoft-owned platform: it killed an AI writing feature it had been promoting for years and replaced it with a tool that lets users snitch on machine-generated dreck. The professional network’s “Enhance Post” button—a one-click AI ghostwriter baked into every text box—is gone. In its place is a new three-dot menu option: “Seems like AI slop.”

The move puts LinkedIn’s billion-plus members on the front line of a content-quality war the platform itself helped start. It also raises urgent questions about who gets caught in the net—and what to do if you’re a genuine professional whose carefully crafted post gets flagged as machine-made.

The Feature That Fueled the Fire Is Now Extinguished

“Enhance Post” wasn’t a minor option. It sat inside the post composer, ready to rewrite your draft into the chummy, bullet-point-dappled corporate tone that became LinkedIn’s unofficial house style. Premium subscribers got even more generative firepower through Microsoft’s Copilot integration. The button was pushed, promoted, and widely used—a direct pipeline for the kind of AI-generated content that, by mid-2026, had turned many users’ feeds into a stream of interchangeable hot takes.

LinkedIn Chief Product Officer Hari Srinivasan announced the retirement on the platform itself, framing it as a return to authenticity. The replacement tool is a narrower proofreader: it fixes grammar and spelling but doesn’t rewrite your voice. “People come to LinkedIn to connect with real people and share their real perspectives, ideas, and expertise,” Srinivasan wrote. “AI slop is a top priority for all of us.”

The timing was not accidental. A study published earlier by Pangram Labs found that more than 40% of LinkedIn’s long-form posts were fully AI-generated—the highest share of any major social platform. An independent analysis by Originality.ai turned up a similar figure. Enhance Post was a major contributor to that statistic, and LinkedIn’s own editorial alarm system was sounding.

How the “Seems Like AI Slop” Button Works

Tapping the new option from a post’s three-dot menu does two things immediately. The flagged content disappears from your feed. More importantly, your report becomes a training signal for LinkedIn’s machine-learning classifiers—the automated systems that decide which posts appear in the recommended feeds of users who don’t follow you.

In plain language: you’re not just cleaning your own timeline. You’re helping LinkedIn’s algorithms learn what “slop” looks like, so they can suppress it across the platform. LinkedIn is separately deploying new classifiers to detect low-quality or AI-generated posts on its own. The human flags are meant to fill in blind spots those classifiers miss.

There’s also a private warning system. Creators whose posts are consistently perceived as inauthentic will get a nudge in their analytics dashboard before any public action is taken. The company says it already blocks “hundreds of thousands” of automated comments every day and has stopped millions of other bot-driven interactions in recent months.

What LinkedIn hasn’t disclosed: how many flags trigger automated suppression, whether flagged creators can appeal, or how the new crowdsourced signals interact with 360Brew, the platform’s existing 150-billion-parameter authenticity model that analyzes lexical diversity and topical consistency.

What It Means for You, Depending on Who You Are

LinkedIn’s change splits neatly across three groups.

Everyday Readers

If you’re tired of scrolling past uncanny-valley career advice, the feed should get incrementally better. The “slop” button gives you a direct way to prune your own experience while feeding the cleanup machine. The risk: false flags. A formal post from a non-native English speaker can read as robotic to an untrained eye. Overuse of the button could inadvertently train classifiers to penalize legitimate voices.

Professionals Who Write and Share Expertise

For IT leaders, recruiters, engineers, and others who post case studies, troubleshooting tips, or deployment lessons, the algorithmic shift could open more oxygen. With less generic content competing for reach, posts grounded in specific, hands-on knowledge should surface more easily. But there’s a catch: if your natural style is concise, data-heavy, or uses predictable structures (as much technical writing does), you could get swept into the slop bucket. This is especially dangerous for writers whose first language isn’t English.

Creators Who’ve Used AI Writing Tools

If you’ve relied on Enhance Post or similar tools to draft content, the message is blunt: stop now. The platform is explicitly drawing a line between “AI as editor” (grammar, spelling) and “AI as ghostwriter.” Posts that smell machine-written will see reduced reach, even if your immediate network engages with them. The private notification system gives you a chance to adjust before your content is publicly suppressed, but that grace period is undefined. The practical advice: switch to the new proofreading tool, write in your own voice, and avoid any tool that rewrites your paragraphs into a generic LinkedIn cadence.

How LinkedIn Painted Itself into This Corner

Microsoft’s AI ambitions made the Enhance Post feature inevitable. Copilot is a headline product across Word, Excel, GitHub, and Bing. LinkedIn Premium tiers bundled AI writing as a selling point. The company trained users to expect AI-assisted posting as a productivity win.

The result was an arms race no one openly wanted. Pangram’s research, which scanned over a million social media posts between April and June 2026, put LinkedIn’s AI-generated post share above 40%—roughly double the cross-platform average. The feed that professionals once used to surface genuine insights became a slurry of recycled thought-leadership fragments.

Substack CEO Chris Best seized on the moment. Nine days before LinkedIn’s announcement, on July 21, Substack integrated Pangram’s AI-detection scanner directly into its reading experience. Best coined “Claudefishing”—a play on catfishing—to describe the practice of passing off AI-generated content as human thought. And he named LinkedIn explicitly as the cautionary tale. The jab stung precisely because it was accurate.

LinkedIn’s response on July 30 was not a technical fix so much as a product-policy line in the sand. By retiring Enhance Post and recruiting users as slop spotters, the platform conceded that AI ghostwriting had become a reach-and-trust liability—even as its parent company continues to sell AI writing tools elsewhere.

The False-Positive Trap Is Real and Documented

Crowdsourced flagging sounds democratic, but it inherits every bias a user brings. A 2023 Stanford study found that seven popular AI detectors misclassified human-written essays by non-native English speakers as AI-generated at a combined false-positive rate of 61.22%. The reason is structural: writers in a second or third language often use more predictable syntax and less colloquial variation—exactly the patterns detection systems associate with machine output.

LinkedIn’s button adds a human layer on top of algorithmic screening. If users disproportionately flag posts that sound formal, unfamiliar, or simply not like the casual-professional argot they expect, those biases get baked into the classifiers the flags are meant to train. LinkedIn has not announced any audit process, no demographic monitoring of flags, and no clear appeal pathway for creators wrongly tagged.

For professionals writing in minority languages, in highly technical fields, or from cultural contexts with different rhetorical norms, the risk is not theoretical. A few misinformed flags could compound into an algorithmic shadowban, and there’s currently no public mechanism to fix it.

What to Do Now: A Practical Checklist

If You’re a Reader

  • Use the “slop” button when a post clearly rehashes generic wisdom with no personal context.
  • Don’t flag posts solely because the style feels rigid or overly polished. If you suspect AI but aren’t sure, consider simply not engaging—your flag is training data.
  • Be mindful that writers whose first language isn’t English may produce prose with patterns that mimic AI. When in doubt, check the author’s profile for human credentials.

If You’re a Creator

  • Abandon AI ghostwriting immediately. Use the new proofreading tool for grammar fixes only.
  • Monitor your analytics for the new inauthenticity flag. If you receive one, review your recent posts for formulaic language or sections that could be mistaken for AI generation.
  • For non-native English speakers: consider adding a brief note in your profile or post that acknowledges your writing style, or deliberately vary sentence structure to reduce false-positive risk.
  • If you’re ever publicly suppressed and believe it’s a mistake, document your original drafts and reach out to LinkedIn support (though an official appeals process isn’t yet confirmed).

If You Manage a Team or Brand

  • Audit your content pipeline. Remove any AI tools that rewrite drafts—the risk now outweighs the convenience.
  • Train your team on the difference between AI editing and AI ghostwriting, using LinkedIn’s new policy as the standard.
  • Watch your company page and employee profiles for unexpected reach declines; cross-reference with the private notification system if it rolls out widely.

What to Watch Next

LinkedIn’s move is a live experiment in platform moderation by crowd. Its success or failure will hinge on whether the classifiers—augmented by human flags—can tell the difference between slop and simply unfamiliar expression. Expect deeper scrutiny in coming months: will false positives concentrate in certain geographies or industries? Will non-native speakers report being penalized? And how will Microsoft square LinkedIn’s anti-slop stance with the aggressive AI-writing push in its other products?

The immediate lesson for every professional is clear: LinkedIn is now a place where authenticity isn’t just praised—it’s algorithmically enforced. Write like a human, because a billion humans now have a button to say you aren’t.