The History of AI

Anthropic is founded

Era 4 · Deep learning and foundation models (2012–2021) › Infrastructure and institutions

Lineage
Machines and computation
Kind
Institutional

Who

Anthropic

What happened

Seven former OpenAI employees, among them the siblings Dario and Daniela Amodei, found Anthropic, an AI safety and research company. In May 2021 it announces a first funding round of $124 million.

Why it mattered

A new frontier lab that presents itself as focused on safety: it sets out to build reliable, interpretable, and steerable systems.

The full story

Anthropic was founded in January 2021 by seven people who had worked at OpenAI, among them the siblings Dario and Daniela Amodei. The company was incorporated as a public benefit corporation, a legal form that requires it to weigh a social purpose alongside profit.

On May 28, 2021, it announced a $124 million round led by Jaan Tallinn, a cofounder of Skype, with participation from Dustin Moskovitz and Eric Schmidt, among others. Dario Amodei was chief executive and Daniela Amodei president. The announcement explained that the team had previously done research on GPT-3, the interpretability of neural networks, scaling laws, AI safety problems, and learning from human preferences.

Its program was to make large AI systems more reliable, interpretable, and steerable, to develop tools for understanding what goes on inside them, and to better integrate human feedback into their development.

In March 2023 Anthropic released its first models, Claude and Claude Instant, and became one of the labs in the race for frontier models.

Builds on

Threads that pass through here

Sources

How to cite

“Anthropic is founded.” The History of AI, revised November 13, 2026. https://thehistoryofai.com/en/milestone/anthropic-founded

This entry was last reviewed on November 13, 2026.

Open in the interactive timeline