Aztec Experiments (github.com)

Aztec has launched a new feature on its v5 mainnet that allows users to post anonymous messages by depositing ETH from Layer 1 (L1) to Layer 2 (L2) without revealing their identity. The system, which includes an anonymous billboard interface, ensures that no sender address is displayed in public call data, maintaining user privacy. Posts can be flagged as "immoral" by a designated censor, who has the ability to enforce moderation through automated mechanisms, including a local LLM daemon that evaluates posts in real-time against an on-chain moderation policy. This development is significant for the AI/ML community as it combines blockchain technology with AI-driven moderation to navigate the nuanced challenges of online speech, privacy, and censorship. The partial formal verification of security properties and the use of cryptographic techniques for anonymity and moderation introduce a layer of robustness that can influence future projects in decentralized applications. The infrastructure enables users to have control over their messages while adhering to a defined moderation policy, balancing freedom of expression with community standards.