GoKawiilIn context: LinkedIn says its "Seems like AI slop" feedback option has been used more than 1 million times since it launched on July 30. The response points to strong user interest in limiting low-quality automated posts, though LinkedIn hasn't said whether the total volume of AI-generated content on the platform is actually falling.
The tool lets members flag posts they believe were generated with AI or appear to be low-effort automated content, and it's available through the three-dot menu on any post. LinkedIn Chief Product Officer Hari Srinivasan said the reports, combined with changes to the company's detection systems, have cut views of content LinkedIn classifies as AI slop by 40% in recent weeks.
LinkedIn is using the feedback as part of a broader effort to curb spam and generic AI-generated material in users' feeds. The company has updated its classifiers, which are systems used to identify content that may have been created by AI, and removed a feature that offered to enhance posts with AI.
It's also rolling out a notice for authors whose posts draw AI-related feedback from other members. The message tells them that some people flagged their post as appearing AI-generated. LinkedIn says the goal is to give authors feedback, not to accuse them of posting spam.
The changes follow growing concern over how much AI-generated content is showing up on professional social platforms. Pangram, an AI-detection company, found that 41% of the LinkedIn longform posts it reviewed were flagged as fully AI-generated.
LinkedIn hasn't detailed how its classifiers work or how it decides when a post gets labeled AI slop. The company appears to be focused on limiting how widely such posts spread rather than banning AI-assisted writing outright, which matters given how many LinkedIn users lean on AI tools to help draft, edit, or organize posts.
Instead, the focus is on content that looks like it was produced and distributed at scale with little human input, things like repetitive posts, generic advice, and automated comments designed to farm engagement.
LinkedIn began moving in this direction earlier in the year, when it said it would crack down on spammy comments produced at scale with little or no human involvement. The latest changes extend that same effort to posts in the main feed.
So far, LinkedIn's early results suggest it's leaning on both user reports and automated detection to limit the reach of suspected AI spam. But the 40% figure only measures fewer views for posts LinkedIn has already classified as AI slop, it doesn't tell us whether AI-generated content on the platform is actually being published less.