Who Owns Claude AI? Anthropic, Investors and Control
If you want to know who owns Claude AI, the answer starts with Anthropic. This artificial intelligence company makes and owns Claude. That includes both the assistant people talk to and the family of AI models that power those conversations.
To understand the full picture, you need to separate ownership, investment and control. Amazon and Google have invested in Anthropic, but giving a company money does not automatically mean owning its products. There are three different questions to consider: Who makes Claude? Who owns shares in Anthropic? And who helps make decisions about how the company is run?
Anthropic Is the Company Behind Claude
Anthropic started in 2021. Its founding team included siblings Dario Amodei and Daniela Amodei. Both had worked at OpenAI, and other former OpenAI employees also helped start Anthropic. The new company became a competitor to OpenAI. It built its own AI models instead of reselling OpenAI's technology.
Dario became the company's chief executive officer, or CEO, while Daniela became its president. These leadership roles are important, but a person's job title does not tell you how much of the company they own. Ownership by founders, authority held by company leaders and oversight by the board are three separate parts of how a company is structured.
Anthropic is a privately owned company, so its shares are not traded on a public stock exchange. Its owners may include founders, employees and outside investors. However, it is not safe to assume exact ownership percentages without reliable public information. Announcements about funding from private companies rarely list every shareholder and show how much of the company each person owns.
The company introduced Claude to the public on March 14, 2023. Claude AI is a product and a family of AI models, not a separate company whose shares are traded on the stock market. Investing in Anthropic therefore means buying an ownership interest in the company behind Claude, not buying a separate part of the assistant itself.
Amazon and Google Hold Investment Interests
Amazon and Google are two of Anthropic's major investors. Their relationships with the company involve more than money: both also provide cloud computing services and help make Anthropic's models available to customers. These partnerships link investment, computing power and access to customers. But they do not make either company the separate owner of Claude.
Amazon first agreed to invest up to $4 billion in Anthropic in September 2023 in exchange for a minority ownership stake. A minority stake means owning less than half of the company. Amazon initially invested $1.25 billion. In March 2024, it added another $2.75 billion, completing its original $4 billion commitment.
In November 2024, Amazon announced another $4 billion investment. This brought its announced total investment to $8 billion at that time. Amazon remained a minority investor. The date is important because a total reported in the past may not reflect the company's latest funding or tell us what percentage of Anthropic Amazon owns.
Google's investment in Anthropic became public in February 2023, along with a partnership for Google Cloud computing services. Google is part of Alphabet. Both Google and Amazon own minority stakes in Anthropic, rather than owning the whole company. Their shares give them an interest in Anthropic's overall business, not separate ownership of the AI models that power Claude.
Venture capital and other funding rounds make the picture more complicated. A company valuation is an estimate of how much a business is worth under the terms of a particular deal. It is not the same as the company's revenue, the amount of cash it has in the bank or proof that an investor controls it. When a company issues new shares, existing shareholders may also end up owning a smaller percentage than before. So the amount of money invested alone cannot tell you who has which voting rights or who controls the company.
How the Anthropic Business Model Works
Anthropic makes money by selling access to its AI technology. Customers can pay to use Claude directly or add it to their own software. This helps explain why people can use the same AI model through different companies without changing who owns it. Anthropic sells access to Claude in several ways:
- Selling subscriptions: Paid Claude subscriptions give people and organizations access to features and usage limits based on their plans. Businesses can also buy plans designed for workplace use. These plans take into account needs such as managing users and meeting an organization's requirements, as well as the answers Claude provides.
- Charging for API usage: An application programming interface, or API, allows software to communicate with Claude. Developers pay according to how much they use it. When their apps send requests to Claude, Anthropic can earn money from that use. Models process text in small pieces called tokens, which are commonly used to calculate charges.
- Offering models through cloud platforms: Customers can use Claude through Amazon Bedrock, which is part of Amazon Web Services, and Vertex AI from Google Cloud. They can also use Anthropic's own services. These platforms give customers more ways to buy and use Claude, but they do not become the owners of the model.
How Anthropic's Governance Shapes Control
Anthropic is registered in Delaware as a public benefit corporation. It is still a business that aims to make a profit, not a charity. Under this legal structure, the company's directors must consider shareholders' financial interests as well as the public benefit the company says it wants to provide and the interests of people who may be significantly affected by its actions.
Corporate governance means the rules and systems used to guide a company and make its leaders answer for their decisions. Anthropic introduced an unusual part of its governance system when it announced the Long-Term Benefit Trust in September 2023. The trust holds a special type of stock that gives it the right to choose and remove members of the company's board of directors.
This arrangement separates some powers over board appointments from ordinary financial ownership. Investing a large amount of money and having the power to appoint directors are not the same thing. To understand who controls Anthropic, you need to look at the different types of shares, voting rights and agreements about how the company is run, rather than simply ranking investors by how much money they have invested.
AI safety also plays a role in Anthropic's stated approach to developing its technology. Its Constitutional AI method uses written principles to guide parts of the process of training models and giving them feedback. This is a technical method, not a rule about who owns the company. Neither this method nor its status as a public benefit corporation guarantees that its AI systems will never make mistakes.
We created this article with the help of AI technology, then checked its facts and edited it with a HowStuffWorks editor.