LinkedIn is rolling out a “seems like AI slop” reporting button, letting users flag posts that appear artificially generated, part of a broader push against low-quality AI content the company’s own chief product officer called “a top priority for all of us.” In the same announcement, LinkedIn confirmed it’s discontinuing its own “enhance your post” AI writing feature entirely, replacing it with a tool that proofreads a user’s existing words rather than rewriting them. A platform building a tool to flag AI content while simultaneously killing its own AI writing feature is what NewsTrackerToday weighs on as the more telling signal than the reporting button itself.
Chief Product Officer Hari Srinivasan was direct about the underlying concern driving both changes: “People come to LinkedIn to connect with real people and share their real perspectives, ideas, and expertise.” That statement frames LinkedIn’s AI slop problem as an authenticity crisis specifically, not simply a content-quality issue, which is why the company is pulling a tool that actively generated writing on users’ behalf rather than just adding better detection on the receiving end.
Isabella Moretti reads the platform-wide pattern this fits into: “Substack added AI-detection tooling through a partnership with Pangram just last week, and Pangram itself raised $9 million in new funding specifically to build AI-content detection at scale, all within the same seven-day window as LinkedIn’s announcement. That’s three separate companies responding to the same underlying problem in the same short stretch of time, which tells you this isn’t LinkedIn reacting to an isolated complaint, it’s an industry-wide inflection point on AI content specifically.” That industry-wide timing, more than LinkedIn’s specific feature, is what NewsTrackerToday checks round as the more significant context behind this rollout.
The scale of the underlying problem is stark by LinkedIn’s own account: the company said it now blocks “hundreds of thousands” of automated comment attempts daily, and millions of other automation attempts in just the past couple of months, a volume that dwarfs what manual moderation alone could plausibly handle without AI-assisted detection tools of its own.
Sophie Leclerc, who covers the technology sector, reads the private-flagging feature as the more thoughtful design choice buried in this announcement: “Rather than only publicly punishing suspected AI content, LinkedIn will privately flag in a user’s own dashboard when their posts read as inauthentic due to heavy AI use, treating it as a nudge to write more like themselves rather than an automatic penalty. That distinction matters because plenty of people use AI tools to polish writing they’ve genuinely composed themselves, and a system that only punishes rather than coaches risks treating legitimate editing assistance the same as fully generated spam.” That coaching-versus-punishing distinction, more than the reporting button, is what News Tracker Today reads round as the more carefully considered part of this rollout.
The reporting button itself is designed to feed LinkedIn’s own detection systems directly, with user reports functioning as training signal to improve the classifiers the company uses to reduce AI slop showing up in suggested content from outside a user’s existing network, meaning the button’s value compounds over time as more people use it.
None of this confirms whether user-driven reporting actually improves detection accuracy meaningfully faster than LinkedIn’s automated classifiers alone would, since crowd-sourced reporting systems on other platforms have had mixed records balancing genuine signal against misuse or false flagging. Whether this combination of reporting, private coaching, and classifier improvements meaningfully reduces the AI slop LinkedIn’s own product chief called a top priority, or whether the underlying arms race between generation tools and detection tools simply continues escalating regardless of any single platform’s countermeasures, is what NewsTrackerToday closes on as the real test this rollout still has to pass.
