SAMPLE DOSSIERJanuary 23, 2026
A

Dario Amodei

CEO & Co-Founder, Anthropic

Princeton PhDEx-OpenAI VP ResearchTIME 100 2025$183B Valuation

1Company Profile

Overview

Anthropic is an AI safety and research company focused on building reliable, interpretable, and steerable AI systems. The company develops the Claude family of large language models and has positioned itself as a leader in responsible AI development.

Company Facts

FoundedJanuary 2021
HeadquartersSan Francisco, CA
StructurePublic Benefit Corporation
Employees~2,300 (Dec 2025)
Valuation$183B (Sep 2025)
ARR$9-10B

Key Investors

Amazon$8B total
Google$3B total
LightspeedSeries E, F lead
ICONIQ, FidelitySeries F
Total Raised$27.3B

Founding Story

Anthropic was founded by seven former OpenAI employees who left due to concerns about AI safety priorities and organizational direction. The founding team included:

  • Dario Amodei (CEO) - Former VP of Research at OpenAI
  • Daniela Amodei (President) - Former VP of Safety and Policy at OpenAI
  • Tom Brown - Lead author on GPT-3 paper
  • Chris Olah - Renowned interpretability researcher
  • Sam McCandlish - Former Head of Research at OpenAI
  • Jared Kaplan - Theoretical physicist from Johns Hopkins
  • Jack Clark - Former Policy Director at OpenAI

Product Portfolio

ModelPricingKey Use Case
Claude Opus 4.5$5/$25 per M tokensFlagship - coding, agents, computer use
Claude Sonnet 4.5$3/$15 per M tokensBalanced performance for coding
Claude Haiku 4.5$1/$5 per M tokensFast, low-latency real-time
Claude Code$500M ARR (2 months)Agentic coding assistant

Competitive Positioning

Anthropic holds 32% of the enterprise LLM market and 41% of AI coding models market share. They are the only frontier model provider available across all three major cloud platforms (AWS, Google Cloud, Azure).

2Executive Profile: Dario Amodei

Personal

Born: 1983, San Francisco
Heritage: Italian-American
Net Worth: ~$3.7B (2025)
Sister: Daniela Amodei (President)

Education

Princeton - PhD Biophysics
Stanford - B.A. Physics (2006)
Caltech - Physics (transferred)
US Physics Olympiad Team (2000)

Recognition

TIME 100 Most Influential (2025)
TIME "Architects of AI" (2025)
Co-invented RLHF
148K+ research citations

Career Trajectory

Co-Founder & CEO at Anthropic
Jan 2021 - Present

Built from 7 to 2,300 employees. Raised $27.3B. $183B valuation.

VP of Research at OpenAI
2016 - Jan 2021

Led GPT-2, GPT-3 development. Co-invented RLHF technique.

Senior Research Scientist at Google Brain
2015 - 2016

Deep learning and neural networks research.

AI Researcher at Baidu
Nov 2014 - Oct 2015

Deep learning research.

Postdoctoral Scholar at Stanford School of Medicine
Dec 2011 - Dec 2014

Neuroscience research on neural circuits.

Why He Left OpenAI

Dario has been explicit about correcting misinformation regarding his departure:

  • Not about Microsoft: Contrary to rumors, his departure was not due to the Microsoft partnership.
  • Vision differences: Found it "incredibly unproductive to try and argue with someone else's vision"
  • Trust concerns: Felt decisions about governance and safety were made without "sincere commitment to safety"

Leadership Style

"Leads through essay-length debates on Slack. Writes long-form essays laying out pros and cons of decisions. This approach gives Anthropic a coherent sense of direction across the entire company — a slower, more reflective counterpoint to Silicon Valley's often chaotic meeting culture."

One colleague noted: "Many people - or really the entire company - have a good model of how he's thinking."

3Organizational Intel

Leadership Team

NameTitleBackground
Dario AmodeiCEO & Co-FounderVP Research at OpenAI; Princeton PhD
Daniela AmodeiPresident & Co-FounderVP Safety & Policy at OpenAI; Stripe
Jared KaplanChief Science OfficerJohns Hopkins professor; theoretical physicist
Sam McCandlishChief Technology OfficerHead of Research at OpenAI; Stanford physics
Ami VoraChief Product OfficerJoined end of 2025; previously Instagram
Mike KriegerLabs LeadCo-founder of Instagram; CTO until 2018
Tom BrownHead of Core ResourcesLead author on GPT-3 paper at OpenAI
Chris OlahResearch DirectorGoogle Brain; interpretability pioneer

Employee Growth

2021 (Founding)
7
2023
240
2024
~1,035+331%
2025
~2,300+122%

Office Locations

San Francisco, CAHeadquarters
Tokyo, JapanFirst Asia office (2025)
Dublin, IrelandEuropean operations
London, UKEuropean operations
Zurich, SwitzerlandResearch hub

Culture & Employee Satisfaction

95%
Would recommend
4.9/5
Culture & Values
4.8/5
Career Opportunities
3.7/5
Work-Life Balance

4Sales Intelligence

Key Insight

Anthropic is in hyper-growth mode with massive infrastructure spending. Decision-making involves technical validation from engineering/research teams. Best entry points: infrastructure scaling, talent acquisition, and enterprise tooling.

Known Vendor Relationships

Cloud Infrastructure
AWS (primary - $8B investment), Google Cloud, Microsoft Azure
Consulting
Accenture (30K professionals trained)
Enterprise Integration
IBM, Snowflake
Legal/IPO
Wilson Sonsini Goodrich & Rosati

Strategic Priorities 2025-2026

Infrastructure ScalingCRITICAL

$13B+ allocated toward compute capacity. 1M+ Trainium2 chips on AWS, up to 1M TPUs on Google Cloud.

Global ExpansionHIGH

Triple international workforce. New offices in EMEA, APAC. Country leads for India, Australia/NZ, Korea, Singapore.

Enterprise GrowthHIGH

Expand 300K+ customer base. Large accounts (>$100K) grew 7x in past year.

IPO PreparationMEDIUM

Hired Wilson Sonsini. Preliminary talks with investment banks. Target: 2026.

Pain Points

Rapid scaling (7 → 2,300) creates operational complexity
Multi-cloud management across AWS, GCP, Azure
Global expansion requiring localized operations
Competing with OpenAI, Google for top AI talent
Enterprise security/compliance certifications
IPO preparation while maintaining 100%+ growth

5Conversation Hooks

Call Prep Summary

Dario values deliberate, research-driven conversations. Reference his recent essays or interviews to show you've done your homework. He leads through written communication - long-form, thoughtful analysis resonates with him. Avoid framing anything as "doomer" — he explicitly pushes back on that label.

Recent Quotes to Reference

"AI is advancing too head-spinningly fast. I believe that these systems could change the world, fundamentally, within two years; in 10 years, all bets are off."

Hard Fork (NYT), February 2025

"I think I'm deeply uncomfortable with these decisions being made by a few companies, by a few people. And this is one reason why I've always advocated for responsible and thoughtful regulation of the technology."

60 Minutes, 2025

"We, as the producers of this technology, have a duty and an obligation to be honest about what is coming. I don't think this is on people's radar."

CBS News

Rapport-Building Topics

The science of scaling — deeply interested in scaling laws and what they reveal
"Machines of Loving Grace" essay — his optimistic vision for AI's positive potential
Building research-first organizations — his approach to Anthropic's culture
Family business dynamics — working with sister Daniela as co-founder
Constitutional AI methodology — the technical approach that differentiates Anthropic
Multi-cloud infrastructure strategy — unique approach to compute diversification

Topics to Avoid

"Doomer" label — he explicitly pushes back on being characterized as a pessimist
Personal attacks on former OpenAI colleagues — discuss vision differences, not people
Father's death — formative experience but deeply personal topic
Suggesting shortcuts in safety processes — company delayed products for safety
Claims that safety focus is "just marketing" — he takes the mission seriously
Microsoft investment in OpenAI — not the reason he left (he's clarified this)

Media Appearances for Deep Research

Lex Fridman Podcast #452
Nov 2024
5+ hours on Claude, AGI, AI safety
Hard Fork (NYT)
Feb 2025
Claude 3.7, AI arms race with China
60 Minutes
2025
AGI development, regulation advocacy
Big Technology Podcast
Jul 2025
AI potential, OpenAI rivalry
The Ezra Klein Show
2024
AI policy and society
Dwarkesh Podcast
2024
Technical AI deep-dive

This Took 1 Minute 47 Seconds

A human researcher would need 3-5 days to compile this depth of intelligence.