LinkedIn, the professional networking platform increasingly known for corporate platitudes and motivational buzzwords, is taking a stand against what many users have long suspected: a significant chunk of the content appearing in their feeds is not written by humans. The company has introduced a new feature that allows members to flag posts as “Seems like AI slop,” giving users a direct way to identify and push back against the flood of automated, formulaic content that has come to dominate the platform.
The announcement came from LinkedIn’s chief product officer, Hari Srinivasan, in a post on the platform itself. Srinivasan explained that the new button is part of a broader effort to improve feed quality and tune the company’s recommendation models. “We are ramping the ability for members to tell us if they believe a post or comment seems like AI slop,” he wrote. “Slop is hard to define and the definition changes; this lets us tune our models and make better feeds.”
The feature works through the three-dot menu located in the upper-right corner of each post. Users who want to report content as AI-generated will simply select the new “Seems like AI slop” option. Once a post is flagged, the person who shared it will receive a private notification in their analytics dashboard, letting them know that their content has been identified as potentially automated. This approach is designed to give feedback to posters without triggering a public takedown or censorship battle, but it does signal a shift in how LinkedIn intends to handle the growing AI content problem.
The Rise of AI Slop on Professional Networks
The term “slop” has evolved to describe the low-quality, mass-produced content that AI tools can generate at scale. On LinkedIn, this often takes the form of over-generalized career advice, inspirational quotes, and hollow leadership lessons that seem to have been written by a robot with an MBA. The problem has become so widespread that LinkedIn users have spent years complaining about the sameness of posts, with some describing the platform as a “cesspool of AI-generated garbage” long before the company officially acknowledged the issue.
According to Srinivasan, LinkedIn has already blocked billions of AI-generated posts and comments in just the last couple of months. That staggering number suggests that automated content had become a significant portion of the platform’s activity. The new button is meant to augment these existing filters by leveraging the collective judgment of LinkedIn’s user base. The idea is that human reviewers can catch things that automated classifiers might miss, especially as AI-generated content becomes more sophisticated and harder to distinguish from human writing.
LinkedIn is not alone in facing this crisis. The broader social media ecosystem has been flooded with AI-generated content, from Facebook pages churning out fake news to X (formerly Twitter) accounts posting automated engagement bait. But LinkedIn is particularly vulnerable because its format encourages professional self-promotion and thought leadership, which often relies on generic, formulaic language. This makes it an ideal breeding ground for AI-generated posts that are designed to appear insightful while requiring minimal effort.
Competitors and the Detection Arms Race
The announcement from LinkedIn came shortly after Substack CEO Chris Best publicly criticized the platform’s reputation for AI slop. In a series of posts on X, Best announced that Substack would integrate an AI detection tool called Pangram into its app. “We built the ability to do a Pangram scan into the Substack app, because we’re sick of slop and we don’t want Substack to turn into LinkedIn,” Best wrote. He added, “Platforms that reward fakeness will drive a race to the bottom. Pangram’s data shows that as much as 40% of long-form text on LinkedIn is generated.”
Pangram, a company that specializes in detecting AI-generated text, published data earlier this month that supported Best’s claim. The company said that LinkedIn had the highest proportion of long-form AI content among all popular social media platforms. “LinkedIn posts made up a third of scanned items, yet it accounted for nearly two-thirds (62%) of all AI content we flagged,” Pangram noted in a blog post. These statistics paint a damning picture of a platform that was once considered the digital home for professional authenticity.
Srinivasan’s response did not attack Substack directly, but it contained a pointed jab at the effectiveness of automated detection tools. “We want members to get feedback from real humans on what sounds authentic – not just have an AI detector review it and get it wrong,” he wrote. This comment highlights the growing debate over whether AI detection software can truly distinguish between human-generated and AI-generated text, especially when the tools themselves are powered by AI.
The Problem with AI Detection Tools
Commercial AI detection tools have become increasingly popular as educators, employers, and platform moderators seek ways to identify automated content. However, these tools are far from foolproof. Studies have consistently shown that they can produce false positives, incorrectly labeling human-written text as AI-generated. This is particularly problematic for non-native English speakers, who may write in ways that mimic the patterns AI models are trained on, such as using more formal or grammatically correct language. Additionally, many detection tools are “black boxes,” meaning the reasoning behind their conclusions is not transparent to the user, making it difficult to know whether to trust their verdict.
The limitations of AI detection are not just a technical issue; they also raise questions about fairness and bias. A human writer who uses em dashes, colons, or other punctuation commonly associated with AI-authored text could be unfairly targeted. Likewise, people who write in a clear, organized style may be flagged simply because AI models are trained on writing that follows proper language rules. As a result, relying solely on automated tools to police content could lead to widespread false accusations and unwarranted penalties.
Humans, for their part, are not much better at identifying AI-generated content than machines. Research has shown that people’s ability to distinguish between AI and human writing is barely better than a coin toss. This is because modern language models have become remarkably adept at mimicking human syntax, tone, and even subtle stylistic quirks. While users may suspect that a post reads as “robotic” or “soulless,” these subjective judgments are often unreliable.
LinkedIn’s Broader Strategy
Alongside the new “Seems like AI slop” button, LinkedIn is ramping up a series of improved classifiers designed to identify AI-generated or generally low-quality content. Srinivasan also announced that the platform would retire its AI-driven “enhance your post” feature, which had previously encouraged users to let AI rewrite their updates. The feature was widely criticized for making posts sound even more generic and robotic. It will be replaced with a more limited AI tool that proofreads writing while maintaining the user’s original voice.
These changes represent a notable shift for LinkedIn, which has increasingly embraced AI in its products and features. The company has integrated AI tools for recruiters, sales professionals, and job seekers, and it has experimented with AI-powered coaching and career advice. However, the backlash against AI slop has forced LinkedIn to balance its enthusiasm for automation with the need to preserve the authenticity that users expect from a professional network.
The success of the new button will depend on how effectively LinkedIn can interpret the feedback it receives. The company says it will use crowd-sourced flags to train its models, meaning that repeated reports of a particular post or type of post could help the algorithm better identify patterns of low-quality content. However, there is a risk that the button could be weaponized by users to suppress content they simply disagree with, rather than content that is genuinely AI-generated. LinkedIn will need to implement safeguards to prevent abuse while ensuring that legitimate flags are taken seriously.
What This Means for Users
For the average LinkedIn user, the new feature offers a way to take action against the deluge of AI-generated posts that clutter their feed. Instead of scrolling past yet another “Unpopular opinion: Hard work beats talent” post, users can now flag it as slop and potentially help improve the platform for everyone. The private notification to the poster is an interesting touch, as it gives the poster the opportunity to reflect on their content and make changes, rather than simply being silently penalized.
However, the notification could also lead to awkward or defensive reactions from users who believe their content is perfectly authentic. Being flagged as an AI slop creator might feel like a public indictment of one’s creativity, even if the notification is private. LinkedIn will need to handle these communications carefully to avoid alienating users who may have unintentionally posted AI-generated content, perhaps by using a third-party tool to draft a post without realizing it.
Ultimately, the fight against AI slop is not just about cleaning up LinkedIn’s feed; it is a broader cultural struggle to preserve the value of human expression in an age of generative AI. As AI tools become more accessible and sophisticated, the line between authentic and automated content will only continue to blur. Platforms like LinkedIn are being forced to decide how much automation they are willing to tolerate before it undermines the trust of their user base.
The company’s promise of “less slop in your LinkedIn feed” is an ambitious one, and it will require a combination of human judgment, algorithmic improvements, and perhaps a healthy dose of user patience. Whether the new button will truly solve the problem remains to be seen, but it is a significant step in acknowledging that the platform’s greatest enemy is not competition, but the endless tide of machine-generated mediocrity.
Source:Gizmodo News