Anthropic’s history from ethical AI startup to global tech powerhouse: the journey from 2021 to 2025

Anthropic was born to make artificial intelligence safer and more independent.

The startup founded by the Amodei siblings has secured billions of dollars from Google, Amazon, and global funds in just a few years, all while maintaining governance focused on ethics and transparency.

From the departure from OpenAI to the birth of a new vision

In January 2021, two siblings—Dario and Daniela Amodei—left the top ranks of OpenAI with a clear idea: to create an artificial intelligence lab that would put safety, model interpretability, and a form of public responsibility at the center. Along with seven other researchers, they founded Anthropic in San Francisco as a public-benefit corporation, a type of company that legally binds its actions to a declared social purpose.

By May of the same year, thanks to the trust of investors like Jaan Tallinn (co-founder of Skype), Dustin Moskovitz (Facebook), and Eric Schmidt (Google), the new company raised $124 million in a first round that allowed it to start training language models focused on transparency and control.

Dario and Daniela Amodei are siblings and co-founders of Anthropic.

Dario, the CEO, holds a Ph.D. in biophysics from Princeton and previously worked at Google Brain and OpenAI, where he led research on large language models. Known for his focus on AI alignment and interpretability, he left OpenAI over concerns about its direction.

Daniela, the President of Anthropic, was formerly VP of People and Safety at OpenAI, overseeing policy and hiring. At Anthropic, she leads governance, operations, and partnerships. Together, they founded the company in 2021 to pursue safer, more transparent AI development.

...FTX’s funds accelerate the development of Claude

In 2022, a huge financial push came from a player who later became a complicated figure: Sam Bankman-Fried, through his trading firm Alameda Research, led a $580 million round. At that time, the FTX empire seemed unstoppable. This influx of capital allowed Anthropic to design and test the first versions of the Claude model, a conversational chatbot alternative to ChatGPT, with a strong emphasis on reliability and the prevention of dangerous behaviors.

The partnership with Alameda would later become controversial after FTX’s collapse, but in the short term, it enabled Anthropic to build one of the most ambitious AI research teams in the world.

Cloud giants compete for influence over the company

In 2023, the major cloud operators started making moves. Google invested $300 million in February, becoming a strategic cloud partner and acquiring about 10% of the company. In May, another round of $450 million arrived, this time with the participation of Salesforce and Zoom.

But it was Amazon that made the boldest move. In September 2023, it announced an investment package of up to $4 billion, starting with an initial tranche of $1.25 billion. The deal required Anthropic to use AWS’s Trainium and Inferentia chips to train its models and to distribute Claude via Amazon Bedrock, Amazon’s AI platform for enterprises.

Governance remains shielded from external pressures

Despite the entry of giants like Amazon and Google, neither obtained seats on the board of directors. Anthropic maintains a hybrid governance structure, with an internal body called the Long-Term Benefit Trust, tasked with overseeing the company’s public mission. This Trust has the power to appoint independent board members and guide corporate decisions toward safety and ethics, even against market pressures.

The consequences of FTX’s collapse and the turning point in 2024

In 2024, with FTX bankrupt, the bankruptcy trustee decided to sell part of its stake in Anthropic. About two-thirds of the shares were sold to a consortium of 24 funds, for a total of $884 million, to help repay some of the creditors of the failed exchange.

Meanwhile, Anthropic kept growing. In February, it closed a new Series D round of $750 million, bringing its valuation to around $18 billion. In March, Amazon completed its initial commitment by paying the remaining $2.75 billion of the mega-deal.

2025 marks an explosion of value and interest

At the beginning of 2025, Google invested another $1 billion, strengthening its strategic position. But it was the March 2025 round that marked the definitive leap: a Series E of $3.5 billion led by Lightspeed Ventures and joined by Fidelity, Cisco, Salesforce, D1 Capital, and other major funds. The post-money valuation soared to $61.5 billion, and Anthropic became the world’s second most valuable AI startup after OpenAI.

By July 2025, rumors began circulating about new negotiations: some investors were reportedly negotiating to buy shares at a valuation exceeding $100 billion, a sign of growing confidence in the strategic value of Claude and the company’s responsible approach.

_______

Anthropic today is a rare example of balance between capital, ethics, and independence.

It has managed to attract record investments from Google, Amazon, and global investors, without ceding control to anyone, building a structure that tries to keep speed, profit, and safety together.

Key figures of the journey

Stage	Funds raised	Main investors	Notes
2021 – Series A	$124M	Tallinn, Moskovitz, Schmidt	Start of research
2022 – Series B	$580M	Alameda/FTX	Birth of Claude
2023 – Google + C	$750M	Google, Salesforce, Zoom	Entry of big tech
2023 – Amazon Bedrock	$1.25–4B	AWS	Preferred cloud partnership
2024 – Series D	$750M	Menlo Ventures	Valuation at $18B
2025 – Series E	$3.5B	Lightspeed, Fidelity, D1	Valuation at $61.5B

_________

FOLLOW US FOR MORE.

DATA STUDIOS

datastudios.org