Decentralized Deliberation for Content Moderation: DAO-Driven Democracy at Scale

Sharma et al. (2024) and Kassen (2023) explore decentralized governance and DAO mechanisms in AI and public sector contexts. However, most platforms still treat moderation as either a corporate or crowdsourced function—rarely as a truly democratic, transparent, and decentralized process. This research would pilot a DAO-based moderation system on a live platform, where content moderation decisions are made via quadratic voting or liquid democracy, incentivizing broad and diverse participation. Rather than a few moderators or opaque AI, the “wisdom of the crowd” becomes proceduralized, with on-chain transparency and auditable decision logs. The project would analyze whether this model increases perceived fairness, reduces bias, and better handles edge cases (e.g., hate speech vs. free expression) compared to both centralized and pure algorithmic approaches. This could radically challenge core assumptions about who should hold moderation power and how legitimacy is constructed in governance.

References:

1. Aligning AI with Public Values: Deliberation and Decision-Making for Governing Multimodal LLMs in Political Video Analysis. Tanusree Sharma, Yujin Potter, Zachary Kilhoffer, Yun Huang, D. Song, Yang Wang (2024). Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society.
2. Blockchain and digital governance: Decentralization of decision making policy. Maxat Kassen (2023). Review of Policy Research.