Edyant is a deep-tech AI governance company founded in 2025.

We benchmark large language models and AI systems across 15 ethical dimensions — including moral reasoning, bias and fairness, safety, privacy, transparency, accountability, cultural diversity, and institutional governance. Our evaluations go beyond capability: we measure how AI reasons, who it impacts, and whether it can be held responsible.

Through Edyant Labs, our open-source research arm, we build and publish frameworks designed to strengthen ethics, transparency, and accountability in real-world AI systems — built to be adopted, audited, and extended by the broader community.

Through GRACE (Global Repository of AI Consensus), we publish a living, versioned AI governance standard built from multi-stakeholder consensus — giving academia, enterprises, and policymakers a shared reference for ethical AI development and deployment.

Through our Consensus platform, we collect statements and positions on AI ethics and governance from individuals, researchers, institutions, and policymakers worldwide — the foundation that shapes and evolves GRACE over time.

We collaborate with academic institutions, enterprises, and policymakers to advance global AI governance. Our work sits at the intersection of technical evaluation, open-source development, and policy — because responsible AI requires all three.

Our name comes from the Sanskrit udyant (उद्यन्त) — meaning "that which rises." We believe responsible AI should rise from human values, not just human capability.

edyant.com

Edyant

Org chart

Teams

Offices