LinkedIn's AI Slop Reporting Button Has Been Used Over a Million Times, Cutting Flagged Content Views by 40%
LinkedIn's "seems like AI slop" reporting button has been used more than a million times since its launch on July 30, 2026, and the Microsoft-owned platform says content it classifies as low-effort AI-generated material is now receiving 40% fewer views than it did a few weeks ago. Chief Product Officer Hari Srinivasan disclosed the figures in an update on the tool's performance. Engadget
The button, unveiled by Bloomberg on July 30, lets users flag posts that appear to be AI-generated filler. LinkedIn has been testing it in recent weeks alongside a set of feed-ranking and policy changes aimed at curbing the volume of AI-generated content on the platform. The company said it uses "many signals" to determine a post's reach and has built safeguards to prevent individual user feedback from being weaponized against other members.
The scale of the problem is measurable. A study by AI detection company Pangram estimated that more than 40% of longform posts on LinkedIn were likely AI-generated. Engadget LinkedIn itself acknowledged a rise in what many call "AI slop" in a May 2026 blog post, while noting that AI can be a helpful tool for refining language. LinkedIn
Alongside the reporting tool, LinkedIn removed its AI-powered "enhance" feature, which let users improve posts and messages using AI. The company had previously made algorithm adjustments to reduce the reach of posts containing common AI phrasing such as "it's not X, it's Y." A new notification, available in the app's post analytics, will alert users when a post they shared was reported by others as seeming like AI slop. Engadget
The anti-slop push sits within a broader set of feed and AI investments. LinkedIn announced in March 2026 that it is upgrading how the Feed ranks content using larger sequence models known as Generative Recommenders, augmented with large language models. LinkedIn The platform is also rolling out the C2PA AI-generated content standard, as announced by LinkedIn VP of Legal Patrick Corrigan. LinkedIn
The C2PA standard, for readers tracking provenance frameworks, is the same content authenticity specification adopted by other major platforms for cryptographic labeling of AI-generated media. Think of it as a digital watermark that verifies where a piece of content came from and whether AI was involved in creating it. LinkedIn pairing a provenance standard with a crowdsourced quality signal is a two-pronged approach: one layer addresses authorship transparency at the content-creation layer, while the other addresses distribution at the feed-ranking layer. The reporting button feeds the latter.
Also this month, LinkedIn announced Hiring Assistant, its first AI agent designed to take on a recruiter's most repetitive tasks. LinkedIn The simultaneous rollout of an AI agent for recruiters and an anti-AI-slop reporting tool captures the tension LinkedIn is navigating: the platform is both building AI-powered products and trying to keep its feed from being drowned by them.
LinkedIn's recent research output frames the environment. A study published August 18, 2026 found women account for just 26% of AI hires as AI jobs surge. The company's 2026 Grad's Guide found 44% of Gen Z respondents cited not having the right network as the biggest barrier to landing an entry-level role. Research published August 4, 2026 found Gen Z professionals in the US are nearly five times more likely than Baby Boomers to fall victim to a job scam. LinkedIn
Additional recent announcements from the platform include the 2026 Top Colleges ranking, the 2026 Cities on the Rise list of 25 emerging US job markets, the expansion of Collaborative Posts to members and Company Pages, Q4 FY26 earnings shared following Microsoft's earnings on July 29, 2026, and the milestone of 100 million members in Brazil, where users created more than 11 million original feed posts and generated 400 million engagement actions. LinkedIn
The broader context here is that platform-level moderation of AI-generated content is still largely uncharted territory. Social platforms spent the last decade building moderation infrastructure for human-produced harmful content: spam, harassment, misinformation. AI-generated low-quality content is a different beast. It is often not rule-violating in any traditional sense, which makes automated takedown inapplicable. The lever platforms have is distribution: demoting content in the feed rather than removing it. LinkedIn's 40% reduction in views for flagged content is an early, concrete data point on how aggressively feed-ranking changes can move the needle.
There is also an open question about false positives. LinkedIn's safeguards against coordinated reporting are described in general terms, and the company has not published detection rates or precision metrics for its AI-slop classification. A user who writes in a style that happens to resemble common AI phrasing could see reduced reach without knowing why. The new post-analytics notification at least gives creators some signal, though it arrives after the fact.
For technology professionals who use LinkedIn as a primary professional channel, the practical takeaway is straightforward: the feed-ranking algorithm now explicitly penalizes content it detects as AI-generated filler, crowdsourced reporting feeds that signal, and a dedicated reporting path exists for users to flag posts. The removal of the "enhance" feature is also worth noting, as it eliminates a first-party tool that may have contributed to the very problem the platform is now combating.
