Characterizing Bluesky Content Moderation Service: From Automation of Service to Landscape of Harms
Updated 1 h ago · first seen 12 Sept 2026
paper_01M29X34XGAM4P4GBD9SBG8X4A
- Published
- 12 Sept 2026
- T1 · 1 h ago
- arXiv
- 2609.11373
- T1 · 1 h ago
- Category
- cs.CY
- T1 · 1 h ago
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.
Authors 8
Abhijnan Chakraborty, Abhisek Dash, Ayan Majumdar, Ingmar Weber, Krishna P. Gummadi, Pushpdeep Singh, Sayeh Jarollahi, Vabuk Pahari
Specification
- Official page
Source:arXiv (Atom API + RSS)T1observed 1 h agohigh
- Arxiv announce type
- cross
Source:arXiv (Atom API + RSS)T1observed 1 h agohigh
- arXiv id
- 2609.11373
Source:arXiv (Atom API + RSS)T1observed 1 h agohigh
- Categories
- cs.AI, cs.CY
Source:arXiv (Atom API + RSS)T1observed 1 h agohigh
Source:arXiv (Atom API + RSS)T1observed 1 h agohigh
- Primary category
- cs.CY
Source:arXiv (Atom API + RSS)T1observed 1 h agohigh
- Published
- 12 Sept 2026
Source:arXiv (Atom API + RSS)T1observed 1 h agohigh
Each value shows its source, tier and observation time. Conflicting claims are kept side by side and flagged — never averaged. How AI Atlas records facts →
Provenance
Attributed facts
9
Source tiers
T19
Freshest observation
1 h ago
Conflicts
None
No models linked to this paper yet.
- Authors
- Pushpdeep Singh, Sayeh Jarollahi, Ayan Majumdar
As of
Rewind the record: see this entity's attributes exactly as AI Atlas knew them on a given day.
Claim history · Abstract
Abstractabstract1
| Value | Valid from → to | Status | Source | Confidence | Extractor |
|---|---|---|---|---|---|
| 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. | → current | current | arXiv (Atom API + RSS)T1 | high | deterministic |
Claims are temporal and append-only: a new observation closes the previous claim (valid_to) instead of overwriting it. Conflicting claims from different sources are kept side by side and flagged — never averaged. Methodology →
- New paperPaperCharacterizing Bluesky Content Moderation Service: From Automation of Service to Landscape of Harms
New paper: Characterizing Bluesky Content Moderation Service: From Automation of Service to Landscape of Harms
arxiv
| Source | Document | Type | Tier | Last observed | Snapshots |
|---|---|---|---|---|---|
| arXiv (Atom API + RSS) | rss.arxiv.org/rss/cs.AI | feed | T1· Official | 1 h ago | 2 |
Tier 1 = official/primary, 2 = quality secondary, 3 = community, 4 = unverified. Every snapshot is archived; see all sources and the methodology.