LinkedIn Takes Steps to Curb AI-Generated Low-Quality Content on Platform

LinkedIn is actively addressing the surge of low-quality, AI-generated content flooding its platform. Last month, the professional network introduced a feature allowing users to report posts and comments suspected of being "AI-generated spam." This move aims to combat the proliferation of inauthentic professional insights and other artificially created content that threatens the platform's credibility.

Unlike social media platforms focused on entertainment and personal connections, LinkedIn positions itself as a valuable source for career information and opportunities for over a billion members. However, AI-generated posts often come across as generic or unauthentic, undermining trust. Brendan Gahan, CEO of Creator Authority, emphasized that while AI is widely used, overreliance can harm credibility, which is crucial for success on LinkedIn.

The challenge of AI content is more significant than previous issues like oversharing, scam ads, and artificial promotion networks. Analysis by Pangram Labs found that between April and June, 41% of long posts and 30% of all public comments on LinkedIn were fully AI-generated, a higher rate than on platforms like X (formerly Twitter) and Reddit. Another startup, Originality AI, identified moderate or higher AI use in 81% of 5,000 public LinkedIn posts sampled in July.

LinkedIn disputes these external figures but acknowledges AI as a useful tool for drafting and refining ideas. The company highlights common AI content markers such as bland conclusions, unusual formatting, repetitive emoji use, and overly polished comments that summarize posts without adding value. To address these, LinkedIn is phasing out its "Improve Your Post" feature, which encouraged AI-assisted editing, and plans to introduce alerts notifying users when their posts are flagged as AI-generated.

Additionally, LinkedIn has enabled hiding comments from unverified users to reduce mass AI-generated replies and blocks over 200,000 AI spam comments daily before publication. The platform claims to detect about 94% of generic or AI-generated posts and limits their distribution beyond the author's immediate network.

LinkedIn's focus is less on whether content is AI-generated and more on ensuring users find valuable, trustworthy information that enhances their professional growth. The growth rate of significantly AI-generated content has slowed recently, rising 15.9% from January to July 2024 compared to 20.3% growth projected for the entire year. However, as AI tools improve, precisely measuring AI content remains challenging. Jeremy Bousino, CEO of Fabicon, noted that no one, including LinkedIn, can accurately quantify AI-generated content volume, and claims otherwise often come from vendors selling AI detection tools.