LinkedIn Says Its New AI Slop Button Has Already Been Used by Over 1 Million People

LinkedIn says its new “Seems like AI slop” button has already been clicked by more than one million people, signaling that the platform’s effort to police low-quality, machine-written posts is getting immediate traction. The company says the feature is helping reduce how much obviously artificial content users see, while also warning posters when their content appears to be generated by AI.

The update matters because LinkedIn has become one of the most visible social platforms where AI-generated writing is spreading quickly. With the company now publicly measuring user reports, classifier improvements, and the reach of suspect posts, it is turning AI moderation into a central product issue rather than a side conversation.

What LinkedIn changed and why it matters

LinkedIn introduced its new reporting option on July 30, adding a “Seems like AI slop” label to the menu that appears when users open the three-dot options on a post. The button gives members a direct way to flag content they believe was created with little human effort, especially the kind of formulaic, over-polished writing that has become common across professional feeds.

Chief product officer Hari Srinivasan said in a Thursday post that the feature has already been used by more than one million members. He added that the company is seeing fewer views on content it classifies as AI-generated, suggesting that the combination of user feedback and ranking changes is beginning to affect distribution.

LinkedIn said it built the feature on the assumption that many people may not be trying to mislead others, but rather are leaning too heavily on generative tools and need a clearer signal that their writing is coming across as machine-made.

The company’s move comes at a time when social platforms are under growing pressure to distinguish human expression from mass-produced AI content. On LinkedIn, that distinction carries particular weight because the site is built around professional credibility, personal branding and workplace expertise. If readers stop trusting what they see, the entire value of the feed weakens.

How did LinkedIn get here?

LinkedIn did not add the AI slop button in a vacuum. The company rolled it out only weeks after outside analysis suggested that a large share of longform posts on the platform appeared to be AI-generated.

According to an analysis by AI detection company Pangram, 41 percent of LinkedIn’s longform posts were flagged as fully AI-generated. That finding, later reported by 404 Media, intensified the sense that LinkedIn’s writing culture was drifting toward templated, synthetic, and repetitive output. The platform’s response was designed to address that perception before it hardened into a permanent reputation problem.

Alongside the reporting tool, LinkedIn said it had improved the classifiers it uses to identify AI content and removed a feature that could “enhance your post” with AI. The company framed those changes as part of a broader effort to keep the platform useful for authentic professional communication rather than promotional automation.

Why LinkedIn is especially vulnerable to AI-generated writing

LinkedIn is more exposed than many other social networks because users are encouraged to publish polished career advice, industry observations, leadership commentary and thought-leadership-style essays. Those formats are easy for large language models to mimic, especially when the output is meant to sound optimistic, authoritative and universally applicable.

That creates a feedback loop. If AI-written posts perform well, more users may be tempted to use AI to produce similar content. Over time, the feed can become crowded with low-variance messages that look credible at a glance but add little original insight. LinkedIn’s labeling and reporting push is an attempt to break that cycle.

What the numbers suggest

The strongest signal so far is not just that users are flagging content, but that they are doing so at scale. More than one million clicks on a moderation tool within a few weeks is a substantial number for a feature targeted at a specific content problem. It suggests that many LinkedIn members are not only noticing AI-style writing, but are motivated to call it out.

LinkedIn also says users are seeing a measurable drop in exposure to content it classifies as AI slop. Srinivasan said the platform is showing “40% less views” on that material than it was a few weeks earlier. That does not mean the content has disappeared, but it does imply that ranking and moderation changes may already be reducing its reach.

Because LinkedIn has not disclosed the exact mechanics behind the classifier or the precise definition of “AI slop,” the metric should be read carefully. Still, the company is clearly signaling that it wants to suppress the spread of material that feels mass-produced, repetitive or inauthentic.

How the new reporting feature works

The AI slop button is tucked into the three-dot menu on a post, which means it is available without interrupting the main browsing experience. Users who believe a post reads like it was generated or heavily polished by AI can tap the option and submit that judgment to LinkedIn.

That structure matters because it lowers the friction of moderation. Rather than forcing users to search for a separate reporting category or fill out a long form, LinkedIn is inviting quick, contextual feedback in the moment a post feels off. In a feed-based product, convenience is often the difference between a rarely used tool and one that becomes part of daily behavior.

What happens after someone flags a post?

LinkedIn has not published a detailed public explanation of every moderation step that follows a report, but the company says the button feeds into its broader detection and ranking systems. In practice, that likely means the platform can use reports as one signal among many when deciding whether content should be labeled, downranked or shown less prominently.

The company’s language also suggests a softer approach than simple removal. Instead of automatically deleting suspicious posts, LinkedIn appears to be using a mix of warnings, classifiers and reduced distribution. That approach gives the platform more room to avoid false positives while still discouraging content that feels automated.

Why LinkedIn is warning posters too

LinkedIn is not only trying to filter what readers see; it is also trying to alert writers when their own posts appear to sound artificial. Srinivasan said users who publish a post may now see a message explaining that “Some members told us this post seems like AI.”

That feedback loop is important because it treats AI detection as a product experience rather than a punishment. A writer who receives that notice may revise the post, change the prompt habits they use with AI tools, or think twice before publishing future content in the same style. In other words, the warning is as much about behavior change as content control.

LinkedIn’s product chief said the company approached the feature with a presumption of good faith, arguing that the goal is to help people understand how their writing may be perceived rather than shame them for using AI tools.

This is a notable choice in an era when many platforms are still figuring out how to handle synthetic content. Rather than making the policy purely adversarial, LinkedIn is trying to steer users toward better judgment and more human-sounding writing.

Why AI slop has become a professional-network problem

AI-generated text is not just a consumer web problem. On professional networks, it can be more damaging because the content is supposed to signal expertise, judgment and real-world experience. If those cues are weakened, users may become skeptical of everything from career advice to hiring commentary.

LinkedIn’s culture also encourages frequent posting, personal branding and evergreen advice. Those incentives can reward content that is easy to generate in bulk: motivational blurbs, workplace aphorisms, and list-driven summaries with little first-hand substance. That makes the platform a natural target for AI writing tools and a natural place for moderation fights to emerge.

• Professional identity: Users expect posts to reflect real expertise.
• Algorithmic incentives: Generic content can still perform well if it is optimized for engagement.
• Scale: AI tools make it easy to publish many polished posts quickly.
• Trust: Overuse of synthetic writing can erode confidence in the platform.

How does this compare with LinkedIn’s earlier AI policies?

LinkedIn has been moving gradually toward tighter AI controls rather than making one dramatic change. Earlier this year, the company said it would crack down on comments created at scale with little or no human involvement. That policy aimed at a related problem: large volumes of repetitive responses that can flood conversations and manipulate visibility.

The new AI slop button extends that strategy to the main feed. Comments are one thing; longform posts are another. By targeting both, LinkedIn is acknowledging that AI-generated material can appear anywhere users interact and that moderation has to cover both original posts and engagement spam.

What changed across the platform

Below is a simplified look at the platform’s recent response to AI-generated content.

What the 40% view drop really means

LinkedIn’s claim that AI-slop content is getting 40% fewer views is the most concrete indication that its interventions may be working, but the number comes with important caveats. The company has not said whether the decline reflects fewer impressions, shorter lifespan in the feed, ranking demotion, or a combination of all three.

Even so, a drop of that magnitude would be meaningful in a network where visibility is the currency that determines whether a post is worth producing in the first place. If content labeled or inferred as AI-generated cannot reliably reach an audience, users may have less incentive to lean on synthetic writing to game engagement.

That said, any moderation system at scale faces the same challenge: bad actors adapt quickly. If AI-generated posts become easier to identify in obvious formats, creators may shift to subtler prompts, mixed human-AI editing, or new stylistic tricks that evade detection while preserving the same underlying churn.

Why the company is framing the issue as “good intent”

LinkedIn’s public messaging matters because it shapes how members interpret the policy. By saying it assumes good intent, the company is drawing a line between people who use AI as a drafting aid and people who flood the platform with low-value content for reach or influence.

That distinction may help LinkedIn avoid alienating users who rely on AI for editing, brainstorming or tone polishing. Many professionals now use generative tools to save time on emails, summaries and first drafts. The issue is not AI assistance itself, but when the final result becomes indistinguishable from disposable machine output.

In that sense, LinkedIn’s policy is less a blanket rejection of AI and more a warning that professional audiences still value originality, specificity and a human point of view. If every post sounds as if it was drafted by the same model, the platform loses one of the things that makes it useful.

What users should expect next

LinkedIn is likely to continue refining the way it detects and labels AI-heavy content, especially if usage of the reporting button remains high. The company has now demonstrated that there is an audience for stronger feed hygiene, and that gives it room to expand moderation tools without appearing out of step with user expectations.

At the same time, the platform will probably keep balancing enforcement with flexibility. Too much false labeling could frustrate legitimate writers, while too little action could let the feed become saturated with formulaic posts. The current approach—reports, classifiers, warnings, and softer ranking changes—suggests LinkedIn is trying to find a middle path.

For now, the takeaway is straightforward: the company’s AI slop crackdown is no longer hypothetical. More than one million people have already used the button, and LinkedIn says the reach of the content it flags is falling. That makes this one of the clearest examples yet of a major platform trying to push back against the industrialization of online writing.

Timeline: how LinkedIn’s AI slop push unfolded

The sequence below shows how quickly the issue moved from concern to product feature.

LinkedIn’s experiment will be closely watched because it sits at the intersection of AI moderation, professional identity and platform trust. If the company can meaningfully reduce synthetic noise without choking off legitimate writing, it may offer a model other social networks eventually follow.