Dario Amodei

Who they are

Dario Amodei is CEO and co-founder of Anthropic — a Princeton-trained biophysicist who left OpenAI's VP of Research role in 2021 to found the company and wrote 'Machines of Loving Grace,' a landmark essay on AI's potential impact on biology, governance, and work.

Person

Dario started at Caltech before transferring to Stanford for his BS in Physics, then took a PhD in Biophysics at Princeton (2011), studying the electrophysiology of neural circuits. He did a postdoc at Stanford's School of Medicine before a brief turn at Google Brain as a Senior Research Scientist, then joined Andrew Ng's team at Baidu as a Research Scientist in 2014. He arrived at OpenAI in 2016, climbed from Research Scientist to Team Lead for AI Safety to VP of Research — the top technical role — spending five years there before leaving in January 2021 to co-found Anthropic. Early in his career he also did part-time software work on Skyline, a mass spectrometry proteomics environment, contributing algorithms and teaching the tool internationally. The through-line is a physicist who moved toward the hardest alignment problems in AI, not away from them. He publishes essays at darioamodei.com — the best-known is 'Machines of Loving Grace,' which lays out a specific, optimistic vision of AI's impact on biology, neuroscience, governance, and work — and has testified before the US Senate Judiciary Committee on AI policy.

Company

Anthropic closed a Series H in May 2026 — $65 billion raised at a $965 billion post-money valuation — surpassing OpenAI (valued at $852 billion in March 2026) to become the world's most valuable private AI company. That followed a $30 billion Series G in February 2026 at $380 billion and a $13 billion Series F in September 2025 at $183 billion — three rounds in under a year. Dario has stated the company could grow by 80 times in 2026 and projected a 130% revenue surge alongside its first operating profit; annualized revenue reached $47 billion as of May 2026. The company has also secured an agreement with Amazon for 5 gigawatts of new compute capacity and partnerships with Google Cloud, Broadcom, and Palantir for enterprise and government deployments. IPO preparations are underway, with Wilson Sonsini advising, though no immediate public listing is confirmed.

Market

Anthropic competes directly with OpenAI in the frontier-model race — OpenAI was valued at $852 billion in March 2026 before Anthropic overtook it. The company is also navigating acute geopolitical pressure: the US government ordered suspension of access to its most powerful models, Fable 5 and Mythos, for all foreign nationals over cybersecurity concerns, a first-ever such directive on a large language model, and the Pentagon has designated Anthropic a supply chain risk — adding regulatory complexity to an already fast-moving competitive environment.

Network

Dario co-founded Anthropic alongside his sister Daniela Amodei, who serves as President and handles operations, partnerships, and external functions — she's his closest operational partner. The founding team includes Ben Mann, Tom Brown, Jared Kaplan, Sam McCandlish, Chris Olah, and Jack Clark — most of them former OpenAI colleagues. He also worked under Andrew Ng at Baidu on Deep Speech, a relationship that shaped his early deep-learning trajectory.

Daniela Amodei· President and co-founder, Anthropic
Chris Olah· Co-founder, Anthropic (interpretability research)
Jack Clark· Co-founder, Anthropic (policy)
Jared Kaplan· Co-founder, Anthropic
Andrew Ng· Worked under at Baidu (Deep Speech)
Ben Mann· Co-founder, Anthropic
Tom Brown· Co-founder, Anthropic

How they likely show up

PhD in Biophysics followed by a postdoc before moving to industry → thinks in first-principles and long research cycles, not quarterly sprints.
Progressed from Research Scientist to VP of Research at OpenAI over five years → earned trust through technical depth before taking on organizational scope.
Left a senior role at OpenAI to found Anthropic on a specific thesis about AI safety → high conviction, willing to take institutional risk on a position others weren't acting on.
Publishes long-form essays (e.g. 'Machines of Loving Grace') rather than frequent short-form posts → processes ideas deeply before going public; prefers substance over volume.
Has testified before the US Senate Judiciary Committee and is active in the Frontier Model Forum → comfortable operating in policy and regulatory arenas, not just technical ones.
Co-founded Anthropic as a Public Benefit Corporation → structures organizational incentives around a stated mission, not just returns; takes governance design seriously.

Conversation tips

→ Reference a specific argument from 'Machines of Loving Grace' — he'll know immediately whether you read it or skimmed a summary.
→ Engage on the tension between the company's safety mission and the Mythos/Fable 5 government access dispute — it's the live version of the questions he's been writing about for years.
→ He came up through physics and biophysics, not CS — analogies from those fields tend to land well.
→ Avoid framing AI safety as a constraint on progress; for him it's the through-line of the work, not a trade-off.
→ Ask about Constitutional AI or interpretability specifically — these are Anthropic's technical bets on alignment and he's been publicly precise about why they matter.

Toolbox

Openers

Open on 'Machines of Loving Grace' — he wrote a specific, named essay at darioamodei.com arguing AI could compress decades of biological and medical progress; it's the clearest window into how he thinks about upside, not just risk.
Reference the Mythos/Fable 5 government shutdown — the US order suspending access to those models for all foreign nationals is the sharpest real-world test of the policy positions he's been articulating since his Senate testimony.
Bring up his early work on the Skyline proteomics software — he contributed algorithms and taught the tool internationally during his PhD years; it's an unexpected through-line between his biophysics roots and his current interest in AI's impact on biology.

Discovery questions

In 'Machines of Loving Grace' you sketched a specific optimistic vision for AI's impact on biology and neuroscience — how has the Mythos launch and the subsequent government access dispute changed how you think about getting those benefits to the people who need them?
You went from VP of Research at OpenAI to founding Anthropic on a specific alignment thesis — what did you see at OpenAI that made you believe a separate institution was necessary rather than pushing from inside?
Claude Code hit $2.5 billion annualized revenue by February 2026 — as a safety-focused lab, how do you think about the tension between an agentic coding product moving that fast and maintaining the interpretability and control properties you've built Constitutional AI around?

Avoid

Don't treat AI safety as a regulatory compliance topic — for Dario it's the founding technical and philosophical premise of Anthropic, and framing it as a box-ticking exercise will end the conversation fast.

Make it yours