LinkedIn Adds a 'Report AI Slop' Button to Fight Junk Posts
LinkedIn is rolling out a user-facing reporting option specifically for low-quality AI-generated content, while simultaneously replacing its own AI writing feature with a proofreading tool. The moves signal a platform-level acknowledgment that AI-generated content has meaningfully degraded the feed.
Original sourceLinkedIn is adding a new reporting option that lets users flag posts as 'seems like AI slop' — a direct admission that the platform's feed has filled up with low-effort, AI-generated content. The feature is part of a broader push to give users more control over what they see and to surface signals the platform can use to demote or filter low-quality posts at scale. It's a notable shift from LinkedIn's previous posture, which leaned into AI-assisted posting rather than pushing back against it.
Alongside the reporting tool, LinkedIn is replacing its own AI writing assistant — which helped users draft posts from scratch — with a proofreading and editing tool. The original feature was widely blamed for contributing to the exact problem the new report button is meant to address: a feed full of identically structured, overly earnest posts that read like they were generated by the same model. Swapping generation for proofreading is an implicit concession that the prior feature made things worse.
The practical impact of the report button will depend almost entirely on what LinkedIn does with the signals it collects. If flagged content feeds into ranking demotion or account-level review, it could meaningfully change posting behavior. If it routes to a moderation queue that buries reports, it becomes a pressure valve with no downstream effect. LinkedIn has not published details on how reported content will be handled algorithmically.
The broader context here is a platform caught between two competing incentives: AI writing tools drive engagement and posting frequency metrics in the short term, while the resulting content quality destroys the feed value that makes the platform worth visiting. LinkedIn is not the first social network to face this tension, but it may be the first to make the trade-off this explicit by literally labeling the problem 'slop' in its own UI.
Panel Takes
The Skeptic
Reality Check
“The report button is only as useful as the moderation pipeline behind it — and LinkedIn has given us exactly zero information about what happens after you click it. Replacing the AI writing feature with a proofreader is the right call, but it's fixing a wound LinkedIn inflicted on its own platform. The cynical read: this is reputation management theater while the algorithmic incentives that reward high-volume AI posting remain completely intact.”
The PM
Product Strategy
“The job-to-be-done here is 'help me see content from real people I trust,' and a report button is a remarkably roundabout way to get there. LinkedIn already has all the behavioral signals it needs to detect AI-generated slop algorithmically — this puts the classification labor on the user instead of solving it in the product. The proofreading pivot is the more interesting decision: it acknowledges that the generation feature was solving the wrong problem, but I'd want to know what the success metric is before calling it a fix.”
The Creator
Content & Design
“The people who write genuinely good LinkedIn posts — specific, personal, earned — were already drowning in a feed full of three-part lists about 'lessons from failure' that no human actually lived through. A report button doesn't fix the incentive structure that rewards volume over voice; it just gives the good writers a way to vent frustration. The proofreading pivot is the right instinct though — helping someone say their thing better is fundamentally different from saying their thing for them.”
The Futurist
Big Picture
“The thesis LinkedIn is implicitly making: human-verified authenticity will become the scarce resource on professional networks, and whoever builds the infrastructure to surface it wins. The dependency that has to hold is that users actually trust LinkedIn's enforcement enough to believe the 'not AI slop' signal — which requires consistent, transparent action on reported content, something LinkedIn has historically been bad at. If this works, the second-order effect is that it creates a template for every content platform facing the same problem: user-generated quality signals as a training layer for feed ranking, which is a more defensible moat than any single content moderation policy.”