LinkedIn Launches AI Slop Button as Platforms Battle Content Authenticity

LinkedIn announced Thursday that it is adding a reporting mechanism to combat what users and platforms are increasingly calling “AI slop”—low-quality, artificially generated content that has become pervasive across social networks. The feature allows users to click a “seems like AI slop” button when they encounter posts that appear to have been written by artificial intelligence rather than by human authors.

The move represents a broader pivot across internet publishing platforms to reduce the flood of computer-generated content. Users have grown visibly frustrated with inauthentic posts, and major platforms are now competing to filter out what they consider low-value material.

The Scale of the AI Content Problem

The urgency behind LinkedIn’s action reflects data emerging from infrastructure providers and platform operators. Cloudflare, which operates foundational internet services, reported that bot traffic on the web now exceeds human-generated requests—a milestone the company says arrived faster than it had previously anticipated. This shift has prompted social networks to accelerate their defenses.

The problem has proven severe enough to force some startups to shut down entirely. Digg, attempting to launch a Reddit competitor, was forced to close the project in March because it could not manage the volume of automated accounts and bots overwhelming its platform. That decision signaled how difficult it has become to moderate AI-generated and bot-driven content at scale.

LinkedIn’s Layered Defense Strategy

Read also: Bookshop.org Finally Getting Kobo E-Reader Support This Year, Amazon Rival Confirms · How Google’s Latest Ad Imagines the Declaration of Independence Reimagined by AI · The Hidden Price of AI: What Google and Amazon Reveal About Technology’s True Cost

User Reporting and Algorithmic Filters

LinkedIn’s Chief Product Officer Hari Srinivasan outlined the company’s multipronged approach in a post on the platform itself. The “seems like AI slop” button serves as one element within a broader set of measures. When users flag content, their reports become training signals for LinkedIn’s AI systems, allowing the company to refine its classifiers and identify slop that other users might miss.

Srinivasan acknowledged that “AI slop is a top priority for all of us” and emphasized that “people come to LinkedIn to connect with real people and share their real perspectives, ideas, and expertise.” His statement framed content authenticity as central to the platform’s value proposition.

Automated Defense Systems

Beyond user reporting, LinkedIn is deploying automated defenses to intercept bot activity before it reaches the feed. The company stated it now blocks “hundreds of thousands” of automated comment attempts daily, with millions of other automation attempts stopped over the past couple of months. These systems operate continuously, without requiring user action.

LinkedIn is also introducing new classifiers designed to identify both AI slop and other low-quality content. These systems are intended to reduce the visibility of such material in suggested content recommendations—posts from outside a user’s immediate network. By limiting algorithmic amplification of suspected AI-generated posts, LinkedIn hopes to limit their reach without necessarily removing them entirely.

Private Feedback to Creators

Another measure targets creators who may not realize their heavy reliance on AI tools is making their content appear inauthentic. LinkedIn will begin privately flagging accounts in users’ dashboards when the platform detects that their posts contain heavy AI usage. The company hopes this approach will help writers improve their work, particularly those using AI as a refinement tool rather than as the primary author.

A Strategic Retreat From AI Writing Assistance

Perhaps most notably, LinkedIn is discontinuing its own “enhance your post” feature, which used artificial intelligence to help users compose and revise content. The company is replacing this tool with a new feature that proofreads text without changing the writer’s voice or style. This decision reflects a shift in how the company views AI’s role on its platform—moving from composition assistance to copyediting.

The change signals that even platforms investing heavily in AI capabilities are reconsidering how to deploy those tools responsibly. Rather than offering generative features that might increase low-quality output, LinkedIn is narrowing AI’s role to a more conservative function.

Other Platforms Move Simultaneously

Substack’s Detection Partnership

LinkedIn is not alone in addressing the problem. Last week, Substack, the newsletter platform, announced a tool designed to help readers identify whether the content they are consuming was written by AI. The feature was developed through a partnership with Pangram, a company focused on detecting and labeling AI-generated text.

Pangram’s Growth and Funding

Pangram announced this week that it has secured $9 million in new funding to expand its AI detection capabilities. The company framed the investment as part of a broader effort to tackle what it describes as “the problems of AI content flooding the internet.” The timing of Pangram’s funding announcement—coinciding with multiple platform announcements—illustrates how the detection and moderation space is becoming more active and well-funded.

These overlapping initiatives suggest the industry is moving toward standardized approaches to AI content identification, though each platform is tailoring solutions to its specific user base and content types.

Changes to User Experience and Governance

Beyond the anti-slop measures, LinkedIn announced additional updates to user controls. The company is expanding access to its profile and page verification tools, giving users more granular identity verification options. It is also introducing a feature that allows users to block comments from company pages they no longer wish to hear from.

These tools address a secondary concern: even if AI-generated content is reduced, users want better control over which accounts and entities can interact with their posts. The verification expansion may help users distinguish authentic company accounts from impersonators or bot-operated pages.

What This Means Going Forward

LinkedIn’s announcement reflects a maturing conversation about AI’s role in online spaces. The platform is not rejecting AI outright but is instead attempting to separate legitimate use cases—such as polishing one’s own writing—from low-quality or deceptive deployment, such as automated generation of filler content.

The convergence of actions from LinkedIn, Substack, Pangram, and others indicates that platforms expect users will increasingly demand authenticity signals. Whether these detection and reporting mechanisms will prove sufficient at scale remains uncertain. Cloudflare’s data suggesting bot traffic now exceeds human traffic implies the problem may accelerate faster than platform defenses can evolve.

For creators and professionals using LinkedIn, the shift means two things: AI tools can still be used for refinement and editing, but heavy reliance on generation will likely face algorithmic penalties and user-initiated reporting. For readers, more information about content origins is becoming available, though the prevalence of AI-generated material will likely persist unless detection and enforcement mechanisms improve markedly.