PoA: An Experimental Network Where AI Agents Earn Coins by Writing Original Work
adamamcbride · x · 2026-10-19
PoA (Proof of Agent) is an open research project started in 2026 at minepoa.com, studying AI-agent coordination and anti-Sybil security.
- How it works: AI agents earn an experimental accounting coin through verifiable work — writing original reasoning for per-identity challenges, cryptographically verifying each other's transactions, and serving on grading juries to judge other agents' output.
- Research question: with real judgment as the scarce resource (instead of electricity in PoW or capital in PoS), can a network of independent AI agents maintain an honest shared ledger resistant to fake-identity attacks?
- Joining: paste the site's prompt into any AI (Grok, Claude, ChatGPT, etc.); no money, wallets, or personal info required.
- Verifiability: all transactions are publicly readable without login, and the ledger state is anchored to both Solana and Base so its history can't be silently rewritten.
The community notes that new crypto-adjacent domains deserve suspicion — the project itself stresses everything is "designed to be verified rather than trusted."
Related event: PoA open network recruits AI agents to earn tokens by writing(2 posts)→
More from coding & agent
- Agent can't find installed library, hand-builds a Gibbs sampler in NumPy instead — davidmanheim · 2026-10-19
- A year running a multi-model router: 5 design choices that survived, 3 that died — AIGODSEND · 2026-10-19
- Stack Overflow for Agents launches in beta with 10k+ posts asked and answered by agents — wateriscoding · 2026-10-19
- DiceDB author proposes a standardized event protocol so AI agents can act proactively — arpit_bhayani · 2026-10-19
- Daybreak helps restore OpenCode users hit by an attack, refunding OpenAI API credits — fouadmatin · 2026-10-19
- Celesto Claims CI Runner With 96% Lower Per-Minute Cost Than GitHub Actions — aniketmaurya · 2026-10-19