LinkedIn adds an “AI slop” report button as platforms rethink AI-written feeds

Lead

LinkedIn is moving more directly against the flood of low-quality AI-generated posts in its feed. According to TechCrunch, the company is adding a “seems like AI slop” reporting option, giving users a way to flag posts that appear machine-written, generic, or inauthentic.

The change is more than a moderation tweak. It shows how a major professional network is recalibrating its relationship with generative AI: useful assistance is still welcome, but mass-produced posts that dilute trust and expertise are becoming a platform problem.

Key points

• A new report signal for AI slop: Users will be able to report posts that seem like low-quality AI-generated content. LinkedIn plans to use those reports as signals to improve its systems for detecting and downranking such material.
• More defenses against automation: LinkedIn says it blocks “hundreds of thousands” of automated comment attempts every day, along with millions of other automation attempts over the past couple of months.
• New classifiers for low-quality content: The company is introducing classifiers to identify AI slop and other low-value posts, especially in suggested content from outside a user’s own network.
• Private feedback for creators: LinkedIn will privately flag in user dashboards when their posts are perceived as inauthentic because of heavy AI use. The goal is to help people improve their writing rather than publicly label them.
• A retreat from AI post-writing: Notably, LinkedIn is removing its “enhance your post” AI feature, which helped rewrite posts. It will replace it with proofreading that corrects text without reshaping the user’s voice.

Why it matters

LinkedIn’s move reflects a broader shift across the web. Platforms once promoted generative AI as a way to increase output and reduce writing friction. But in social and publishing environments, more content does not automatically mean better content. If feeds become filled with polished but empty AI-written posts, users may lose confidence that they are hearing from real professionals with real experience.

TechCrunch also notes similar pressure elsewhere. Substack recently added a tool, through a partnership with Pangram, to help users identify AI-written content. Pangram has also raised new funding to address the spread of AI-generated material online. At the infrastructure level, Cloudflare has said bot traffic now exceeds human-generated requests on the web, underscoring how automation is reshaping online activity.

LinkedIn’s approach is relatively careful. It is not banning all AI-assisted writing. Instead, it is trying to distinguish between light editing or proofreading and content that appears to be fully outsourced to a machine. That distinction will matter as more platforms develop policies around AI: the real question is not only whether text was generated by AI, but whether it is useful, authentic, and aligned with user expectations.

For professionals and creators, the message is clear. Template-like AI posts and automated engagement are likely to become less effective on LinkedIn. Posts grounded in lived experience, subject-matter expertise, and a recognizable human voice will matter more as the platform tries to restore signal in the feed.

Source: TechCrunch AI