Report: AI-Generated 'Slop' Floods LinkedIn and Social Media Platforms

A new report from AI-detection firm Pangram highlights a significant surge in AI-generated content across social media, with LinkedIn emerging as a primary source of what researchers term as AI slop. By analyzing over one million posts across platforms including X, Reddit, and LinkedIn, the firm found that 13.8% of scanned content was fully AI-generated. LinkedIn accounted for 62% of all flagged AI content, despite representing only a third of the sample. Long-form posts on the platform were particularly prone to automation, with over 40% of posts exceeding 250 words being identified as machine-authored.

The rise of AI slop is fueled by the ease of generating long-form business advice, leadership content, and marketing copy through LLMs. Pangram’s data suggests that the longer the post, the higher the probability of AI usage, reaching 25.72% for content over 250 words across all platforms. Comments on LinkedIn are also increasingly likely to be AI-generated, further diluting the platform’s focus on human networking and professional engagement. This proliferation presents a challenge for content moderation teams attempting to distinguish authentic human contributions from repetitive, algorithmic output.

LinkedIn has acknowledged the prevalence of repetitive, low-quality automated content and is actively adjusting its recommendation algorithms to prioritize human-authored material. The company claims its internal detection systems effectively identify generic, repetitive content with high accuracy. While the current study serves as a convenience sample and not a comprehensive dataset, it underlines a growing sentiment among users to curb the influx of synthetic material. Moving forward, the effectiveness of platform-led moderation and user-led AI detection will be critical in preserving the utility of professional networking environments against synthetic content proliferation.