LinkedIn rolled out a new feature this week to curb AI slop on the platform; users were first to point out.

A new option now sits under the three-dot menu on any post to flag anything that “Seems like AI slop.” Once clicked, the platform responds with “Thanks for letting us know. Your feedback helps improve the feed,” and then hides the post.

This morning, LinkedIn’s chief product officer Hari Srinivasan confirmed the changes in a post , acknowledging that AI slop is “a top priority” for the company and outlining its efforts to combat it.

Srinivasan said the button will feed into LinkedIn’s models directly, which is useful because “slop is hard to define and the definition changes.” LinkedIn says it has also continued to invest in automated detection, noting the platform flags hundreds of thousands of automated comments attempts every day, and that they have blocked “billions of other automation attempts” in the last couple months alone.

LinkedIn will also privately test flagging when other members feel a post came off as inauthentic or overly AI-generated in a user’s own analytics dashboard. Srinivasan stressed the importance of having LinkedIn’s users determine what feels authentic and what doesn’t, and not leaving it up to an AI detector that “gets it wrong.”

LinkedIn has been scrutinized as a breeding ground for AI slop, both by the media and frustrated users. Those AI detectors have been used to confirm what many users suspect: as much as 41% of long-form written content seen by users is likely to be fully AI-generated, Pangram data revealed in early July.

Substack took a dig at the platform last week when announcing its own AI detection feature ; CEO and cofounder Chris Best said, “we don’t want to wait until your Substack app turns into LinkedIn before we start to learn and make progress.”

So far, users appear to be celebrating the latest feature. “Hopefully this is the first step in making the platform more enjoyable for everyone, and reduces the content we all hate,” The Story Club founder Jonny Rose wrote in a LinkedIn post . Others, like tech ghostwriter Colin Steele, pointed out that the button could be “ripe for abuse,” where companies and its employees report competitors without merit.

LinkedIn did not explicitly say what happens when a user’s post is flagged as AI-generated, other than that it's used to help improve its feed. At this point, it's unclear whether reports could limit the reach of certain posts or accounts.

The update comes a few weeks after Reddit reinforced its position on AI slop following reports that AI-generated responses could be manipulated with as little as a 13-word snippet of text on the platform.

But as with Reddit, the bigger question is how any person or detection system can say with complete certainty that a piece of text was generated by AI.

Most people can scroll LinkedIn and spot writing that's blatantly machine-generated. But proving it beyond doubt is a different story—especially as users avoid clichéd sentence structures and other tells commonly associated with AI writing.