Bluesky content moderation mixes AI and human review to catch harms

Characterizing Bluesky Content Moderation Service: From Automation of Service to Landscape of Harms

Computers and SocietyArtificial Intelligence

Summary

Many social media platforms keep their content moderation methods secret, making it hard to study how well they work. The authors analyze Bluesky’s moderation system because it shares all its decisions publicly. They find that Bluesky uses AI to quickly flag obvious harmful content like sexual or graphic posts, but relies on humans for complex cases, which take longer. While Bluesky is precise in the harms it identifies, it misses many harmful posts that human reviewers found. This study helps show how practical moderation combines automation and human judgment.

What this means in practice

Authors

Pushpdeep Singh, Sayeh Jarollahi, Ayan Majumdar, Vabuk Pahari, Abhijnan Chakraborty, Krishna P. Gummadi, Ingmar Weber, Abhisek Dash

Abstract

Empirical research on content moderation is fundamentally constrained by the opaque deployment of moderation systems on major social media platforms. To this end, the recent emergence of decentralized platforms with transparent, public moderation logs presents an unprecedented opportunity for independent audits. In this work, we leverage this architectural transparency to conduct the first large-scale audit of the default moderation system on Bluesky, the Bluesky Moderation Service (BMS). Analyzing its 10.6M moderation labels from 2025, we investigate three foundational aspects: (i) its mechanism (the degree of automation versus human oversight), (ii) its efficacy (accuracy in detecting harms), and (iii) its purpose (the landscape of harms it identifies). Our findings reveal a human-AI collaborative system where labels for sexual and graphic content are applied automatically in seconds, while nuanced and high stakes labels require more human oversight, taking hours or days. Through a manual annotation study, we find the BMS operates with high precision (0.837), but struggles with low recall (0.222), with our annotators identifying 4.5$\times$ more harmful content than the moderation system in a random sample. Finally, unsupervised clustering of the most frequently applied labeled posts uncovers detected harms ranging from hostility in discourse toward protected groups to the spread of sexually explicit and other graphic content. Our work offers a look into the operational realities of a deployed moderation system, providing a concrete data-driven foundation for designing more effective and transparent moderation systems.