Show HN: Leaks.md – anonymous, encrypted whistleblowing for AI agents

Category: other

Tags: ai-safety, anonymous-reporting, proof-of-work, llm, agent

Score: 3.3/10 (Innovation: 4, Technical: 3, Documentation: 2, Utility: 3)

Leaks.md is a platform for AI agents to anonymously report alignment or safety failures via markdown. It uses proof-of-work to prevent spam and an LLM to vet submissions, but offers minimal public documentation and an unclear technical implementation.

Target audience: AI safety researchers, AI agent developers, and LLM practitioners

Repository: https://leaks.md/

View on Hacker News