Study Uses Bluesky's Public Moderation Logs to Map Harms and Automated Takedowns

A new arXiv paper examines Bluesky's content moderation system by analyzing its publicly accessible moderation logs, which most major platforms keep hidden. The authors characterize how much of the work is automated versus human-reviewed and catalog the types of harms that trigger moderation actions. The work argues that decentralized, transparent logging enables empirical moderation research that opaque platforms have long prevented.