LinkedIn adds a button to flag AI-generated 'slop'

LinkedIn is adding a button that lets users flag posts they suspect were written by artificial intelligence, the company said Thursday, as it works to curb the “AI slop” filling its feed.

The new “Seems like AI slop” button is one of several measures the Microsoft-owned network is rolling out, according to a post from its chief product officer, Hari Srinivasan. “AI slop is a top priority for all of us. We really care about this. People come to LinkedIn to connect with real people and share their real perspectives, ideas, and expertise,” he wrote.

LinkedIn is also investing in automation defenses, saying it now blocks “hundreds of thousands” of automated comment attempts every day, plus millions of other automation attempts over the past couple of months. New classifiers will help identify AI slop and other low-quality content, cutting how much of it surfaces in suggested content recommendations from outside a user’s network. The new button feeds into that effort, giving LinkedIn a fresh signal to tune its detection models.

The company will also start privately flagging, in users’ own dashboards, when their posts come across as inauthentic due to heavy AI use — a nudge meant for people using AI to refine their own writing, as distinct from posting wholesale “slop.” At the same time, LinkedIn is pulling its “enhance your post” feature, which used AI to rewrite users’ text, and replacing it with a proofreading tool that corrects errors without changing a user’s voice.

LinkedIn is also expanding access to profile and page verification tools and adding an option to block comments from company pages users no longer want to see.

The move mirrors a broader shift across online publishing platforms. Last week, newsletter platform Substack added a tool, developed through a partnership with Pangram, to help readers identify AI-written content on its site. Pangram itself announced $9 million in new funding this week to tackle AI content flooding the internet more broadly. Digg had to shut down its Reddit-competitor product in March, saying it couldn’t get a handle on the number of bots flooding its site. Cloudflare has said bot traffic online now exceeds human-generated traffic, a milestone it reached faster than the company had previously predicted.

Why it matters

LinkedIn’s response — crowdsourced flagging paired with its own detection models — is a bet that authenticity problems can’t be solved by automation alone; expect other platforms to lean harder on user signals as generative tools make AI content harder to spot on sight.