Aurelius

Aurelius is a decentralized alignment protocol that surfaces and evaluates reasoning within AI systems. It uses a multi-agent architecture of Miners, Validators, and the Tribunate to generate adversarial prompts, evaluate ethical reasoning, and synthesize collective judgment for verifiable AI alignment.