Anthropic

Holds a BA in Literature (minor in Politics) from UC Santa Cruz. Joined Stripe as one of its earliest employees, initially in recruiting before moving into risk management — a non-engineering path that shaped her operational approach. Spent roughly three years at OpenAI as VP of Safety and Policy, overseeing trust and safety work. Co-founded Anthropic in 2021 alongside her brother Dario and five other departing OpenAI researchers, betting that safety and capabilities could be built together rather than bolted on after the fact. As President, she runs the business and policy side of Anthropic while her brother leads research and product. She has been notably public about keeping Anthropic structured as a public-benefit corporation and consistently frames AI safety as a governance challenge, not just a technical one — a posture that reflects her unusual route from an English degree and Stripe ops into the center of frontier AI.

Earned a PhD in biophysics from Princeton as a Hertz Fellow, followed by a postdoc at Stanford School of Medicine, before shifting from neuroscience to AI. Was a senior research scientist at Google Brain, then joined OpenAI in 2016 as a safety team lead, eventually serving as VP of Research where he led the development of GPT-2 and GPT-3 and co-invented reinforcement learning from human feedback (RLHF). In 2021, co-founded Anthropic with his sister Daniela to pursue AI safety from outside the OpenAI結構. A distinguishing public posture: he has said he spends 30–40% of his time on company culture rather than products, arguing that culture — not technology alone — will be the decisive competitive advantage in the AI race.

Started as a technical journalist — covering distributed systems at The Register and becoming Bloomberg's "world's only neural network reporter" — before joining OpenAI in 2016 as an early policy and communications hire. He rose to Policy Director over four years, then left to co-found Anthropic. Throughout that transition, he has maintained **Import AI**, a widely-read weekly newsletter analyzing AI research and policy that he launched around the time he joined OpenAI. He also co-chairs Stanford HAI's AI Index, sits on the National AI Advisory Committee, and is a non-resident fellow at CSET. His journalism-to-practitioner arc gives him an unusual vantage: he reads, summarizes, and contextualizes frontier research at a pace that mirrors his own newsletter's cadence, and his public writing often doubles as an informal record of how the field's norms are shifting.