LinkedIn has evolved into a digital hub where professionals connect, recruiters hunt for talent, and would-be thought leaders broadcast their insights to the world. Alongside legitimate career updates and industry analysis, however, a new kind of content has flooded the platform: AI-generated posts that are often formulaic, low-effort, and strikingly similar in tone. These posts, commonly referred to as "AI slop," have become a source of irritation for many users. In response, LinkedIn is testing and rolling out a new reporting feature: a "Seems like AI slop" button. The company hopes this will help curb the spread of content that feels disingenuous, while giving users a direct way to shape the quality of their feeds.

The announcement came from Hari Srinivasan, LinkedIn's chief product officer, who detailed the company's latest efforts to combat AI-generated content in a post on the platform. He acknowledged that not all AI-generated content is necessarily bad, as some users rely on AI tools to refine their ideas or improve their writing. The problem, he noted, is the vast quantity of low-quality, mass-produced posts that are published solely to game algorithms and attract engagement. LinkedIn's new button is designed to identify those posts more effectively by combining user feedback with automated detection systems.

Key facts at a glance

• LinkedIn is introducing a "Seems like AI slop" button for users to report posts that appear to be low-quality AI-generated content.
• The company is also deploying new automated classifiers to detect AI slop, especially in comments.
• Users who are flagged will be able to see private feedback on their dashboards when their audience deems posts inauthentic.
• Reported posts will have reduced reach, and users should see fewer such posts in their feeds.
• A study by Originality.ai found that 81% of long-form posts on LinkedIn are likely AI-generated.

What exactly is AI slop?

The term "AI slop" has gained enough traction to earn a dictionary entry. Merriam-Webster defines it as "digital content of low quality that is produced usually in quantity by means of artificial intelligence." The phrase has become a catch-all for the kind of content that fills social media feeds with repetitive, shallow, and sometimes nonsensical texts or images. The Associated Press has also added an entry to its style guide, underscoring how widespread the phenomenon has become.

AI slop can appear in many forms: mass-produced AI-generated images, music, videos, and text posts. On LinkedIn, the typical AI slop post follows a recognizable template. It might begin with a line like "Most executive lessons aren't learned in the boardroom" before pivoting to an unexpected setting, such as a coffee shop or a soccer game. The post is often broken up into short, punchy sentences with dramatic line breaks. It references "today's fast-paced world" or "the future of leadership." It ends with an invitation to "let that sink in." And despite the confident tone, the post usually amounts to very little actual insight.

These posts are not limited to LinkedIn. From Facebook to X (formerly Twitter) to Substack, users have noticed a surge in AI-generated content that seems designed to perform well in algorithms rather than provide value. Some AI slop is harmless, but much of it is created to generate views, clicks, and even revenue. The phrase "Fruit Love Island," for example, refers to a bizarre AI-generated image trend that circulated on Facebook, showcasing fruit engaging in human-like activities. That content might be strange but relatively innocuous. The concern grows when AI slop spreads misinformation or drowns out genuine voices.

How LinkedIn's new button will work

According to Srinivasan's post, LinkedIn has been investing in automated defenses to identify and remove low-quality content. The platform is introducing new classifiers that can spot patterns commonly associated with AI slop. These models will work alongside user reports to flag suspicious posts.

The "Seems like AI slop" button will appear as an option when users choose to report a post. When a user selects this option, LinkedIn will take the feedback into account. If a post receives enough reports, it may be shown to fewer people or removed entirely if it violates the platform's policies. The system will also give posters a private dashboard notification when audience members feel their content is inauthentic. This feedback loop is intended to nudge creators toward more genuine, higher-quality posts.

This approach is notable because it relies on the judgment of the community as much as on automated detection. AI detection tools can be imperfect, and some AI-generated content may be entirely legitimate. By allowing users to make the call, LinkedIn hopes to strike a balance between removing spam and respecting those who use AI responsibly. For example, a user might use an AI writing assistant to outline a post, then rewrite it in their own voice. That post would not be considered slop, because it contains genuine thought and effort. The new system is designed to catch posts that are mass-produced, generic, and lacking original insight.

Why LinkedIn is a magnet for AI slop

LinkedIn's culture has long rewarded a certain kind of performative posting. "Hustle culture" and "thought leadership" are celebrated, and users are encouraged to share their professional wisdom on a regular basis. This creates an environment where consistency and engagement are prized. Hootsuite notes that LinkedIn's algorithm rewards topic relevance, engagement, and consistency. A user who posts multiple times a day with AI-generated content about "C-suite leadership" may see more reach than someone who posts thoughtful, original content a few times a month. The incentive structure thus favors quantity over quality, which is exactly what AI slop thrives on.

The problem is compounded by AI-generated comments. Some users employ bots or AI tools to leave generic comments like "Great insights!" or "Thanks for sharing this!" on posts, artificially inflating engagement metrics. This makes it even harder for authentic conversations to surface. LinkedIn's new classifiers are also aimed at identifying these low-quality comments, which can skew the platform's understanding of what content is valuable.

Srinivasan acknowledged that AI-generated content is not always problematic. People use AI in different ways: to draft emails, to brainstorm, or to polish their writing. The distinction, he said, lies in whether the content is created in good faith and adds value, or whether it is mindless and disingenuous. By giving users a way to signal the latter, LinkedIn hopes to reduce the noise without penalizing all AI-assisted creativity.

Broader industry efforts against AI slop

LinkedIn is not alone in tackling this issue. Substack, the email newsletter platform, has announced a partnership with Panagram, an AI detection company, to help readers understand how much of a post may be AI-generated. This transparency tool is aimed at giving readers more context about the content they consume. Reddit communities, meanwhile, have become ground zero for complaints about AI slop. Subreddits dedicated to books, hobbies, and even niche sports have reported an influx of strange, AI-generated text posts that seem to have no purpose other than to farm engagement.

The rise of AI slop is closely tied to the public launch of ChatGPT in late 2022 and its rapid adoption in 2023. The ability to generate human-like text at scale opened the door for content farms and individual users to flood the internet with cheap, mass-produced articles and posts. What started as an exciting technological capability has turned into a content quality crisis. Platforms are now struggling to keep up, and users are becoming more savvy at spotting AI-generated content.

The "Seems like AI slop" button is a small but significant step. It represents a shift in how platforms think about moderation: instead of relying solely on AI, they are turning to the community for help. It also signals that the initial enthusiasm for generative AI is being tempered by a more realistic understanding of its downsides. Low-quality AI content may not disappear entirely, but with tools like this, it could become less visible.

For now, LinkedIn users can look forward to a less cluttered feed, at least when it comes to the most egregious examples of AI slop. The company's move could inspire other platforms to develop similar features, as the battle between AI-generated content and human authenticity continues to evolve.

Source: ZDNET News