LinkedIn adds a 'seems like AI slop' button so users can flag fake-sounding posts
The Microsoft-owned professional network is blocking hundreds of thousands of automated comments a day and scrapping its own AI writing tool, because its members want to hear from real people, not bots.
Key points
- LinkedIn launched a "Seems like AI slop" report button on Thursday, letting members flag posts that appear computer-written rather than genuine.
- The platform blocks hundreds of thousands of automated comment attempts daily, and millions of other bot actions over the past two months.
- Substack added an AI-content label tool last week through a partnership with detection company Pangram.
- Pangram raised $9 million in new funding this week to help platforms fight low-quality AI content.
- Cloudflare says bot traffic now outpaces human web requests, a threshold it didn't expect to reach this soon.
LinkedIn has a slop problem. Chief Product Officer Hari Srinivasan said this week that AI slop, meaning posts mass-produced by AI tools rather than written by a real person drawing on real experience, has become one of the platform's biggest member complaints. "People come to LinkedIn to connect with real people and share their real perspectives and expertise," he wrote.
The new report button is the most visible fix. Click it on any post that feels hollow or machine-generated and LinkedIn logs that signal, feeding it into new classifiers, software filters that scan posts and score them on authenticity, to reduce how often low-quality AI content appears in recommendations beyond your own network.
What else is LinkedIn actually doing?
The button is one piece of a wider crackdown. LinkedIn now blocks hundreds of thousands of automated comment attempts every day. It's also rolling out private dashboard warnings: if enough people flag your posts as inauthentic, you'll get a quiet heads-up rather than a public shaming.
One move stands out. LinkedIn is pulling its own "enhance your post" feature, which used AI to rewrite drafts entirely, replacing it with a tool that proofreads your words but keeps your voice intact. The company helped create the very problem it's now fighting.
Is this happening everywhere, not just LinkedIn?
Far beyond one platform. The same week LinkedIn made its announcement, Substack added a labelling tool so readers can see when a newsletter was AI-generated, built with Pangram, a startup that detects AI-written text. We covered both moves as they broke: the Substack tool on 22 July and Pangram's $9 million raise on 29 July.
Cloudflare, which handles traffic routing for a large part of the web, reports that bot-generated requests now outnumber requests made by actual humans, a threshold that arrived faster than its own forecasts suggested. Digg tried to rebuild a Reddit-style community this year and shut down in March; founders said bots flooded the platform faster than they could remove them.
| Platform | Action | When |
|---|---|---|
| "Seems like AI slop" report button plus classifier filters | This week | |
| Pulled AI post-rewriting tool | This week | |
| Substack | AI-content labels via Pangram partnership | Last week |
| Pangram | $9 million funding round | This week |
| Digg | Shut down Reddit competitor due to bot overload | March 2025 |
What should you do if you use LinkedIn for work?
Don't panic about using AI to help you write. LinkedIn's private warning system targets posts that read as fully machine-generated, not posts where you used a tool to tighten a sentence you already wrote. The proofreading tool they're launching is built precisely for that lighter use.
Use AI to sharpen your thinking, not to replace it. A post that starts with your genuine opinion and gets tidied up with a tool reads differently from one generated out of a blank prompt. Your LinkedIn network can tell the difference. Apparently, so can LinkedIn's new software.
Watch whether the classifier actually changes what appears in your feed, not just what gets reported. That's where this either works or becomes another moderation checkbox that nobody notices.
