Dario Amodei

With an education rooted in both physics, which delivers an understanding of complex modelling systems, and biophysics, with knowledge about the mechanics of human neural circuits – it’s no surprise that Dario is at the forefront of AI technology.

As the former vice president of research at OpenAI, and now CEO and founder of Anthropic, he is almost better known for leaving a company than creating one. But with Anthropic now valued at over $183 billion, the story of how it began is fast-being eclipsed by what it has become.

Born in 1983 in San Francisco, Dario grew up in the Mission District of the city and discovered his passion for maths and numbers at an early age. At high school he loved physics and maths and went on to study physics at both Caltech and Stanford. Having finished his degree at Stanford, he went to Princeton where he pivoted to biophysics, studying under Professor Michael Berry, whose primary research is in neural computation. Dario really wanted to understand how the brain works.

After graduation, Dario joined Baidu’s Silicon Valley AI Lab, where he worked under Andrew Ng and supported the development of Deep Speech 2, a ground breaking end-to-end neural network that inspired deep learning voice systems. He did a short stint at Google before leaving to join Open AI – where the founding team was looking for deep-learning talent.

At OpenAI, Dario quickly made his mark, rising to the position of Vice President of Research. During his tenure, he supervised significant projects, including the development of GPT-2 and GPT-3, and contributed to the advancement of reinforcement learning from human feedback (RLHF) techniques. His sister Daniela joined in 2018 as engineering manager.

After four and a half years at Open AI, Dario and Daniela left the business – citing a difference in vision for the company. Dario was keen to drive a more responsible approach to AI implementation, and to prove that this would be ethical and profitable. In 2021, they founded Anthropic, with seven others – all but two of which had also left Open AI.

Established as a public benefit corporation, Anthropic focuses on building reliable, interpretable, and steerable AI systems. Under his leadership, the company has developed – Claude, a series of large language models, emphasising safety and alignment through approaches like "Constitutional AI” – aligning AI with human values. Anthropic has attracted substantial investments, raising over $33B of capital since inception.

Dario is recognised for his contributions to AI policy and safety discussions. In July 2023, he testified before a U.S. Senate judiciary panel, warning of the potential dangers of AI, including its use in weaponry. 
In 2024, Dario published an essay titled, “Machines of Loving Grace,” an optimistic vision for how AI could change our lives, in response to people’s assumption that he is a “doomer” who just thinks about the risk with AI. He wrote: “One of my main reasons for focusing on risks is that they’re the only thing standing between us and what I see as a fundamentally positive future.”