Who owns Anthropic? Ownership structure explained (2026)

If you've used Claude, the AI assistant built by Anthropic, you've probably wondered who actually owns the company behind it. That question has become more pressing as Anthropic has grown into the most valuable AI startup in the world. Its last primary funding round, in May 2026, valued it at $965 billion, and its shares changed hands near a $1.2 trillion valuation on secondary markets by mid-2026.

Ownership matters here more than it does for most startups. Anthropic sits at the center of a global race to build powerful AI systems, and the companies funding that race — Amazon, Google, and others — have their own strategic interests in the outcome. Who controls Anthropic shapes how its AI models are developed, how safety research is prioritized, and how the technology gets distributed across the economy.

This article breaks down Anthropic's ownership structure, its key investors, the people who run the company, and the unusual governance mechanisms designed to keep it focused on its safety mission.

Anthropic is an AI safety and research company best known for building Claude, a family of large language models. The company was founded in 2021 by Dario Amodei (CEO) and Daniela Amodei (President), along with several other former members of OpenAI's research team. It is headquartered in San Francisco.

The company's stated mission is to build AI systems that are safe, beneficial, and understandable. Its primary products include the Claude 4.5 model family (led by Claude Opus 4.5 and Claude Sonnet 4.5), an API for developers, agentic tools such as Claude Code and Cowork, and a consumer-facing chat interface. Anthropic has also published influential safety research, including work on constitutional AI and model interpretability.

Anthropic's run-rate revenue crossed $47 billion in May 2026, up from roughly $1 billion at the start of 2025, driven by API usage, enterprise contracts, and its Claude Code coding tool. Its May 2026 Series H round valued the company at $965 billion post-money, and secondary trades have since implied a valuation above $1 trillion. Anthropic employs roughly 5,000 people, with plans to nearly double headcount by the end of 2026, and competes directly with OpenAI, Google DeepMind, Meta AI, and xAI in the foundation model market.

Because Anthropic is privately held, it doesn't publish a public shareholder registry. Ownership is distributed among founders, employees (via equity grants), and a growing list of venture capital and strategic investors. No single entity holds a majority stake.

What makes Anthropic's structure unusual is its corporate governance layer. The company is organized as a public benefit corporation (PBC), a legal designation that requires directors to balance shareholder returns with a stated public benefit — in this case, AI safety. On top of that, Anthropic has established a Long-Term Benefit Trust (LTBT), an independent body designed to hold a special class of shares that can override certain decisions if the company drifts from its safety mission.

This structure is intentional. It's meant to prevent any single investor — including Amazon or Google — from steering the company's research agenda purely for commercial gain.

Anthropic has raised more than $100 billion in total funding across a rapid succession of rounds. Here's a breakdown of the most significant capital injections:

Amazon's investment is the largest by a wide margin. Its commitment has grown well beyond the initial $1.25 billion made in September 2023 and the $2.75 billion added in March 2024, with a further $5 billion committed in 2026 tied to a new multi-gigawatt compute agreement. In exchange, Anthropic uses Amazon Web Services (AWS) as its primary cloud and training partner and makes Claude models available through Amazon Bedrock, AWS's managed AI service.

Google has invested billions of dollars since 2023 and, in October 2025, expanded the partnership with a deal giving Anthropic access to as many as one million of its TPU chips, worth tens of billions of dollars. In November 2025, Microsoft and Nvidia joined the cap table, committing up to $5 billion and up to $10 billion respectively, and Claude became available on Microsoft Azure. Claude is now the first frontier model offered across all three major clouds: AWS, Google Cloud, and Azure.

Dario and Daniela Amodei, along with other co-founders, hold significant equity stakes, though exact percentages are not publicly disclosed. As with most venture-backed startups, early employees also hold stock options or restricted stock units that represent a meaningful share of the company's cap table.

Given the volume of outside capital raised, founder dilution is a real factor. However, the LTBT structure and Anthropic's PBC status give the founding team governance protections that go beyond simple equity math. Control, in Anthropic's case, is not purely a function of who owns the most shares.

In June 2026, Anthropic confidentially filed an IPO prospectus with the SEC, its clearest step yet toward a public listing. A confidential filing lets a company begin the SEC review process without immediately publishing its financials, and it does not lock in a timeline. With run-rate revenue above $47 billion and reports of its first quarterly profit approaching, Anthropic sits firmly in IPO-ready territory, though no listing date has been set.

Understanding who owns Anthropic requires distinguishing between economic ownership (who holds equity) and operational control (who makes decisions day to day).

Dario Amodei co-founded Anthropic in 2021 after serving as VP of Research at OpenAI. He holds a PhD in computational neuroscience from Princeton. As CEO, he sets the company's research direction, oversees product strategy, and serves as its most visible public spokesperson. His public statements consistently emphasize safety-first development, positioning Anthropic as a counterweight to what he describes as reckless scaling by competitors.

Daniela Amodei serves as President, overseeing business operations, go-to-market strategy, sales, and policy. Before Anthropic, she held leadership roles at OpenAI and Stripe. Her operational role complements Dario's research focus, and together the siblings form the company's core leadership team.

Anthropic's board includes representatives from its major investors, though the company has been deliberate about limiting investor influence over research decisions. The Long-Term Benefit Trust is designed to act as an independent check, a body that can intervene if the company's leadership or investors push it away from its safety commitments. In 2026, former Federal Reserve chair Ben Bernanke was appointed to the trust, adding a high-profile outside voice to that oversight layer.

Amazon, despite being the largest financial backer, does not have a board seat that grants it control over Anthropic's strategic direction. This was a negotiated condition of the investment, and it distinguishes Anthropic's relationship with Amazon from, say, Microsoft's deeper integration with OpenAI.

Anthropic's ownership story is compressed into just five years, but it covers a remarkable amount of ground — from a small research lab to one of the most valuable private companies in tech.

The speed of Anthropic's valuation growth is striking. In roughly five years, the company went from a $5 billion valuation to $965 billion in its last primary round, and to about $1.2 trillion on secondary markets. That is close to a 200-fold increase, driven by explosive demand for foundation models and a historic AI infrastructure buildout.

Anthropic's ownership structure has attracted attention from regulators, particularly around the Amazon and Google investments.

Both the U.S. Federal Trade Commission (FTC) and the UK's Competition and Markets Authority (CMA) have examined whether large strategic investments by Amazon and Google in AI companies like Anthropic constitute de facto acquisitions. The concern is that these investments — while structured as minority stakes — could give big tech companies outsized influence over the AI market without triggering formal merger review.

In 2024, the FTC opened an inquiry into the competitive dynamics of major AI partnerships, including Amazon-Anthropic and Microsoft-OpenAI. The CMA conducted a similar review. Neither regulator blocked the investments, but both signaled ongoing interest in monitoring how these relationships evolve.

Anthropic's Long-Term Benefit Trust is a novel structure with no direct precedent in Silicon Valley. It's designed to prevent mission drift, but it has also raised questions: Who sits on the trust? How are its members selected? Under what circumstances would it actually intervene?

These details remain partially opaque. Critics argue that the LTBT is untested and could prove toothless if the company faces pressure to maximize returns ahead of an IPO. Supporters counter that its mere existence creates a structural check that most AI companies lack entirely.

Who owns Anthropic isn't just a corporate trivia question. It has direct implications for several things you likely care about.

Product direction: Amazon's investment means Claude is deeply integrated into AWS. Google's backing ensures availability on Google Cloud. These partnerships shape where and how you can access Anthropic's models — and which platforms benefit most from them.

Safety priorities: Anthropic's PBC status and LTBT are designed to keep safety research central to the company's mission. If ownership shifted toward investors with purely commercial incentives, that balance could change.

Competition: The concentration of AI investment among a handful of tech giants — Amazon, Google, Microsoft — raises questions about whether smaller players can compete. Anthropic's ability to remain independent, despite taking billions from two of the largest cloud providers, is a test case for the entire industry.

Pricing and access: As Anthropic scales its enterprise business, its ownership structure will influence how it prices API access, whether it favors certain platforms, and how aggressively it pursues revenue versus broad availability.

Dario Amodei is the CEO of Anthropic. He co-founded the company in 2021 after serving as VP of Research at OpenAI. He holds a PhD in computational neuroscience from Princeton and is responsible for the company's overall research direction and strategy.

Not yet. Anthropic remains a privately held company as of mid-2026, with no stock ticker and no shares on any public exchange. It did confidentially file an IPO prospectus with the SEC in June 2026, an early step toward a potential public listing, but no date has been announced.

Anthropic was founded in 2021 by Dario Amodei and Daniela Amodei, along with approximately ten other former OpenAI researchers. The founding team left OpenAI over differences in approach to AI safety and commercialization.

Amazon is the largest single investor, having committed more than $8 billion. Google has invested billions more and holds a significant minority stake, while Microsoft and Nvidia joined in late 2025 with commitments of up to $5 billion and up to $10 billion. Other major backers include ICONIQ, Lightspeed Venture Partners, Sequoia Capital, Altimeter, Salesforce Ventures, Spark Capital, and Menlo Ventures. The Amodei siblings and other co-founders hold meaningful equity stakes, though exact percentages are not public.

The Long-Term Benefit Trust (LTBT) is an independent governance body that holds a special class of Anthropic shares. It's designed to ensure the company stays aligned with its AI safety mission, even under pressure from investors or in the event of an IPO. The trust can override certain corporate decisions if it determines the company is drifting from its stated purpose.

Both companies were founded with safety-oriented missions, but their structures differ. OpenAI is controlled by a nonprofit foundation that oversees a for-profit public benefit corporation, a structure it finalized in 2025. Anthropic is a public benefit corporation with the LTBT as an additional governance layer. Anthropic's largest backer, Amazon, does not have board control, whereas Microsoft's earlier and much larger investment in OpenAI came with closer operational integration.