Dario Amodei

Dario Amodei is co-founder and chief executive of Anthropic, the AI safety-focused company behind the Claude family of large language models. A physicist by training, he became one of the most prominent researchers in machine learning before leaving OpenAI, where he served as vice president of research, to found Anthropic in 2021 alongside his sister Daniela Amodei and several other former OpenAI colleagues.

Anthropic was founded on the premise that increasingly powerful AI systems require rigorous safety research to avoid catastrophic risk, a philosophy Amodei has articulated in widely read essays such as 'Machines of Loving Grace' and in his warnings about the pace of AI capability growth. Under his leadership the company has grown into one of the best-funded AI labs in the world, raising capital from investors including Google and Amazon and reaching a valuation in the tens of billions of dollars.

Before Anthropic, Amodei held senior research roles at Google Brain and Baidu, where he worked alongside Andrew Ng on deep learning for speech recognition. At OpenAI he helped lead development of GPT-2 and GPT-3, work that shaped his views on the scaling laws that now underpin the entire generative AI industry, and on the safety measures needed to accompany them.

In 2026 Amodei has become an increasingly visible figure in AI governance, briefing the United Nations Security Council on AI risk, meeting with US President Donald Trump at the White House, and publicly clashing with parts of the administration over what he has called AI 'doomerism' framing. He has consistently argued for a middle path between unchecked acceleration and outright bans, advocating for transparency, interpretability research and targeted regulation.

Amodei holds a PhD in biophysics from Princeton University and has said his scientific background shapes his approach to treating AI development as an empirical, safety-first discipline. He remains one of the industry's most influential advocates for taking AI's transformative and destructive potential equally seriously.