LinkedIn Launches 'AI Slop' Reporting Button Amid Platform-Wide Authenticity Push

LinkedIn is deploying algorithmic and user-driven defenses against synthetic content, introducing a flagging mechanism for posts suspected of being AI-generated while simultaneously retraining its recommendation models to de-amplify such material. According to The Verge AI, the company’s chief product officer Hari Srinivasan characterized the effort as a response to user demand, stating that “AI slop is a top priority for all of us.”

The ‘AI Slop’ Reporting Mechanism

The new button appears in the three-dot menu on individual posts and allows users to report content that “seems like AI slop.” When activated, LinkedIn hides the flagged post and confirms receipt of the feedback. According to The Verge AI, Srinivasan indicated the company will use this signal to “tune our models and make better feeds,” suggesting the reports feed directly into content-ranking retraining pipelines rather than serving as simple volume counters.

The Scale of AI-Generated Content

The timing of LinkedIn’s enforcement push reflects data on the prevalence of synthetic content on the platform. According to 404Media, AI detector Pangram found that 41% of longform LinkedIn posts were flagged as completely AI-generated. This figure provides a concrete baseline against which to measure the effectiveness of LinkedIn’s new classifiers—systems the company says will identify posts that are “AI-slop or generally low-quality content” and reduce their distribution.

Architectural Changes to AI-Assisted Writing

Beyond detection and reporting, LinkedIn is reframing its relationship with generative AI in post composition. According to The Verge AI, the company is retiring a feature that used AI to “enhance” posts—language implying content rewriting—and replacing it with a proofreading tool designed to “not change your voice.” This shift signals a distinction between voice-preserving assistance (positioning as legitimate) and voice-altering synthesis (positioning as inauthentic), a distinction that may also inform how the new classifiers evaluate user-generated content.

Why This Matters

LinkedIn’s multi-layered approach—user reporting, algorithmic detection, feed demotion, and feature redesign—reflects the platform’s recognition that authenticity is now a competitive differentiator in professional social networks. For professional users, the reporting button provides agency in shaping feed quality; for the platform, user-generated labels create training data for classifiers and surface which accounts or topics drive the highest concentration of flagged synthetic content. The deprecation of the AI-enhancement tool suggests LinkedIn is shifting its value proposition from “AI helps you write better” to “AI helps you write clearly without losing your voice”—a narrower, less controversial positioning. If the classifiers prove effective and the 41% baseline drops materially, other platforms may adopt similar detection-and-demotion strategies. Conversely, if bad-faith actors game the reporting mechanism or the classifiers suffer high false-positive rates, the approach could amplify human-vs.-synthetic tensions without reducing actual spam or deceptive content.