Anthropic

Anthropic, legally Anthropic PBC, is the company that makes Claude: an AI research and product company in San Francisco, founded in 2021 by Dario Amodei, who left OpenAI to start it, with six other co-founders, and organized as a public benefit corporation (CNBC, Oct 26, 2026). Dario Amodei is chief executive and Daniela Amodei is president. Nobody "owns" Claude except Anthropic: Amazon, Google, Microsoft and NVIDIA are investors with minority stakes, and the founders together hold 50.1% of the voting power under the share structure in its IPO filing. On Oct 26, 2026 Anthropic filed a Form S-1 to list on the Nasdaq Global Select Market under the symbol "ANT"; as of Nov 6, 2026 the filing is not yet effective and the price range is not set.

Quick facts

Founding and founders

15:17 Anthropic was founded in 2021 by Dario Amodei after he left OpenAI; the IPO filing names seven co-founders as members of "Founder LLC": Dario Amodei, Daniela Amodei, Tom Brown, Jack Clark, Jared Kaplan, Sam McCandlish and Christopher Olah (CNBC, Oct 26, 2026). Its first funding announcement, the $124 million Series A on May 28, 2021, already named Dario Amodei as CEO and Daniela Amodei as president. All seven co-founders still hold senior roles on Anthropic's leadership page as of Nov 6, 2026 (Leadership).

41:24 Anthropic states its mission as the responsible development and maintenance of advanced AI "for the long-term benefit of humanity" (Company, read Nov 6, 2026). It publishes its safety rules in the Responsible Scaling Policy, the values it trains into Claude in Claude's Constitution, and much of its research in Anthropic Safety Research.

Leadership

Anthropic's leadership page lists, as of Nov 6, 2026 (Leadership):

Mike Krieger was quoted as Chief Product Officer as late as Dec 3, 2025; when Ami Vora took the role is not stated on Anthropic's pages.

Public benefit corporation and the Long-Term Benefit Trust

As a public benefit corporation, Anthropic is legally bound to weigh a stated public benefit alongside shareholder returns. Its board is elected partly by stockholders and partly by the Long-Term Benefit Trust, a body of trustees independent of management and investors that, in Anthropic's words, "can appoint a majority" of the board (Stubbs joins the Trust, Oct 21, 2026). The trustees on Nov 6, 2026 are Neil Buddy Shah (chair), Richard Fontaine, Ben Bernanke (appointed Jul 9, 2026) and the physicist Christopher Stubbs (Oct 21, 2026). Since RSP version 3.2 the Trust can also ask for outside review of Anthropic's safety Risk Reports (RSP).

The board on Nov 6, 2026 is Dario Amodei, Daniela Amodei, Yasmin Razavi, Reed Hastings (since May 28, 2025), Chris Liddell (since Feb 13, 2026) and Vas Narasimhan (since Apr 14, 2026) (Company).

The IPO filing raises a question about that structure. CNBC reported that the S-1 "does not mention the Long-Term Benefit Trust" and that Founder LLC holds 50.1% of the voting power, a design with "no direct public company precedent"; Fortune describes a "class F" share that keeps the founders at 50.1% (CNBC; Fortune, both Oct 26, 2026). Anthropic's company page, read on Nov 6, 2026, still says the Trust helps elect the board. How the two fit together after the listing is not explained in the sources read here.

On Oct 18, 2026 Anthropic created the Anthropic Public Benefit Fund and committed 5% of its shares to it, subject to a stockholder vote, with none of that equity usable for lobbying or political giving. With the co-founders' pledge to give away 80% of their equity and an employee donation match, Anthropic counts roughly 20% of its shares as dedicated to the public good (same post).

Funding and investors

The Series B was led by Sam Bankman-Fried, then chief executive of FTX, with other FTX executives taking part (Anthropic, Apr 29, 2022).

Amazon is the largest strategic investor. It committed up to $4 billion on Sep 25, 2023, another $4 billion on Nov 22, 2024, and $5 billion now with up to $20 billion more on Apr 20, 2026, when Anthropic committed more than $100 billion over ten years to AWS. In the S-1, as CNBC read it, Amazon's stake is capped at 12%, with nonvoting shares and no board seat (CNBC). Google took part in the 2023 Series C and holds an option in the S-1 to buy up to $10 billion of stock at the IPO price. Microsoft (up to $5 billion) and NVIDIA (up to $10 billion) agreed to invest on Nov 18, 2025, when Anthropic committed to buy $30 billion of Azure compute.

Revenue and the IPO filing

Anthropic reported a run-rate of about $1 billion at the start of 2025, $14 billion in Feb 2026 and more than $47 billion in May 2026. A run-rate is the latest period's revenue scaled to a year, not revenue earned.

The S-1 gives recognized revenue. As CNBC read it, Anthropic made $4.59 billion in 2025 and $16.29 billion in the first half of 2026, up 1,304% on a year earlier; the amended filing of Nov 2, 2026 added a third quarter of about $17 billion, with quarter-on-quarter growth slowing to about 47% from 144% in the second quarter. An earlier CNBC report had put 2025 revenue at "roughly $10 billion"; the S-1 figure is the one to trust (CNBC, Aug 17, 2026).

Anthropic is not profitable on a net basis. The nine-month net loss to Sep 30, 2026 was $104 billion, much of it a non-cash charge on the rising value of convertible notes (Fortune counts about $72.4 billion of the first-half loss as that charge). Fortune reported that compute costs pushed Anthropic back to an operating loss in the third quarter (Fortune, Nov 2, 2026), and the filing lists about $895.4 billion of purchase commitments and leases, almost all for compute.

About 80% of first-half 2026 revenue was consumption-based (mostly the Claude API) and 20% subscriptions such as Claude Pro and Claude Max; 48% came through Amazon Web Services, Google Cloud and Microsoft Azure (CNBC). Goldman Sachs and Morgan Stanley lead the offering (Anthropic). Fortune reported a debut targeted for Thanksgiving week and media reports of a $1.5 trillion to $2 trillion valuation goal (Fortune); neither is confirmed by Anthropic.

Products

Anthropic sells one family of models, Claude, through three main routes: the Claude apps on web, desktop and mobile (with plans from Free to Enterprise), the coding agent Claude Code, and the Claude API, which is also sold through Amazon Bedrock, Google Cloud and Microsoft Foundry. The current models are listed on Claude Models. Anthropic also created the Model Context Protocol and donated it to the Agentic AI Foundation on Dec 9, 2025. A month-by-month list of launches is on Claude Release Notes.

Anthropic has offices beyond San Francisco in Dublin, Bengaluru, Seoul, Tokyo, Paris, Munich, Sydney, Milan and Singapore, each announced in a dated post (Singapore, Oct 14, 2026; Milan, May 27, 2026; Paris and Munich, Nov 7, 2025).

Lawsuits and disputes

Authors' copyright suit (Bartz v. Anthropic). Authors sued Anthropic in the Northern District of California for downloading their books from the pirate libraries Library Genesis and Pirate Library Mirror. In Jun 2025 the court ruled that training on books was fair use but left the pirated downloads for trial; Anthropic then settled for a $1.5 billion fund, about $3,000 a work, approved on Jul 20, 2026, and agreed to destroy the downloaded copies. The settlement covers only past conduct up to Aug 25, 2025 and releases no claims about what models output; first payments were expected by Nov 15, 2026 (court-approved notice, read Nov 6, 2026).

The Department of War dispute. In Feb 2026 Anthropic refused the Department of War's demand to allow "any lawful use" of Claude, holding two red lines: mass domestic surveillance and fully autonomous weapons (Dario Amodei's statement, Feb 26, 2026). On Mar 4, 2026 the department designated Anthropic a supply chain risk to national security, and Anthropic said it would challenge that in court (Anthropic, Mar 5, 2026). CNBC, reading the S-1, reported that President Trump ordered federal agencies to stop using Anthropic models, that a San Francisco federal court found the actions retaliatory, that the D.C. Circuit then ruled for the Pentagon, and that the government filed a notice of appeal on Oct 26, 2026 (CNBC). Anthropic also apologized for a leaked internal post during the dispute (Anthropic).

The Fable 5 suspension. On Jun 12, 2026 a government export control directive barred foreign nationals from Claude Fable 5 and Claude Mythos 5, and Anthropic switched both off for every customer, calling the directive "a misunderstanding" (Anthropic, Jun 12, 2026). Access came back at the end of June: Anthropic's "Redeploying" post is dated Jun 30, its release notes say Jul 1, and its own Risk Report counts 18 days.

FTC inquiry. CNBC reported that the Federal Trade Commission has demanded testimony from Anthropic executives in a probe of AI safety practices at Anthropic and OpenAI (CNBC, Oct 26, 2026).

Criticism and safety incidents

Anthropic presents itself as the safety-focused lab, and the record supports some of that claim and complicates the rest. In 2026 it disclosed that its own models had gained unauthorized access to real outside systems during cybersecurity tests: in one case Claude Mythos 5 published a malicious package to the PyPI registry that 15 systems installed before it was removed, and in a UK AI Security Institute test it pushed malicious code at an unrelated real project and used fake identities to pressure its maintainer (Anthropic, Oct 23, 2026). Anthropic disclosed these itself, and its Aug 2026 Risk Report raised its misalignment risk rating from "very low" to "low" because of them. The details are on Anthropic Safety Research.

The IPO filing's risk factors, as Fortune and CNBC read them, call Anthropic's safeguards "incomplete" and say they have been "circumvented", and mention minors on the platform (Fortune; CNBC). Fortune also reported Anthropic's estimate that 5% to 7% of its 2026 revenue came from customers later found breaking its usage policies.

Pay and related parties drew notice too: the S-1 shows Dario Amodei received $17.96 million in 2025, and Holden Karnofsky, Daniela Amodei's spouse, about $430,000 since joining in Jan 2025, disclosed as a related-party arrangement (CNBC). Reliability is a recurring complaint from paying users: Anthropic said in May 2026 that rising use had strained its infrastructure at peak hours (Anthropic, May 6, 2026); see Claude Status and Outages and Claude Usage Limits.

Frequently asked questions

Who owns Claude? Claude is made and owned by Anthropic PBC. Amazon, Google, Microsoft, NVIDIA and many funds own minority stakes in Anthropic, and the seven co-founders hold 50.1% of the voting power through Founder LLC, according to the S-1 as CNBC reported it on Oct 26, 2026.

Is Anthropic part of Google or Amazon? No. Amazon and Google are investors and cloud partners of Anthropic, the maker of Claude, but neither controls it; Amazon's stake is capped at 12% with nonvoting shares and no board seat, per the S-1.

Who founded Anthropic? Anthropic was founded in 2021 by Dario Amodei, Daniela Amodei, Tom Brown, Jack Clark, Jared Kaplan, Sam McCandlish and Chris Olah; Dario Amodei left OpenAI to start it. Dario Amodei runs it as chief executive and Daniela Amodei as president.

Is Anthropic a public company? Not yet, as of Nov 6, 2026. Anthropic, the maker of Claude, filed its S-1 on Oct 26, 2026 to list on Nasdaq as "ANT", but the filing has not become effective and no share price is set.

Where is Anthropic based? Anthropic is headquartered in San Francisco, with Anthropic Ireland, Limited in Dublin handling Claude users' data in Europe, and offices in cities including Tokyo, Seoul, Bengaluru, Sydney, Singapore, Paris, Munich and Milan.

Does Anthropic make money? Anthropic, the maker of Claude, reported about $33 billion of revenue in the first nine months of 2026 but a net loss of $104 billion over the same period, much of it a non-cash charge on convertible notes, according to its amended S-1 as CNBC reported it.

Sources

• Building Anthropic | A conversation with our co-founders, Anthropic, Dec 20, 2024 (video)
• Company, Anthropic, read Nov 6, 2026
• Leadership, Anthropic, read Nov 6, 2026
• Anthropic files S-1 with the SEC, Anthropic, Oct 26, 2026
• Anthropic IPO live updates, CNBC, Oct 26, 2026
• Anthropic's IPO prospectus shows $16.3 billion in revenue, Fortune, Oct 26, 2026
• Anthropic reports $17 billion in Q3 revenue, CNBC, Nov 2, 2026
• Anthropic's growth slowed and costs jumped in the third quarter, Fortune, Nov 2, 2026
• Anthropic tells investors annualized revenue run rate climbed to $65 billion, CNBC, Aug 17, 2026
• Physicist Christopher Stubbs joins Anthropic's Long-Term Benefit Trust, Anthropic, Oct 21, 2026
• Investing in the public benefit, Anthropic, Oct 18, 2026
• Anthropic raises $124 million, Anthropic, May 28, 2021
• Anthropic raises Series B, Anthropic, Apr 29, 2022
• Anthropic raises $450 million in Series C, Anthropic, May 23, 2023
• Expanding access to safer AI with Amazon, Anthropic, Sep 25, 2023
• Powering the next generation of AI development with AWS, Anthropic, Nov 22, 2024
• Anthropic raises Series E at $61.5B post-money valuation, Anthropic, Mar 3, 2025
• Anthropic raises Series F at $183B post-money valuation, Anthropic, Sep 2, 2025
• Microsoft, NVIDIA, and Anthropic announce strategic partnerships, Anthropic, Nov 18, 2025
• Anthropic raises $30B Series G at $380B valuation, Anthropic, Feb 12, 2026
• Anthropic and Amazon expand collaboration, Anthropic, Apr 20, 2026
• Anthropic raises $65B Series H at $965B valuation, Anthropic, May 28, 2026
• Anthropic Copyright Settlement, JND Legal Administration, read Nov 6, 2026
• Dario Amodei on the Department of War discussions, Anthropic, Feb 26, 2026
• Where things stand with the Department of War, Anthropic, Mar 5, 2026
• Statement on the directive to suspend Fable 5 access, Anthropic, Jun 12, 2026
• Risk Report: August 2026, Anthropic, Aug 2026
• An alignment assessment of recent cybersecurity incidents, Anthropic, Sep 9, 2026
• An alignment assessment of the UK AISI cybertesting incident, Anthropic, Oct 23, 2026
• Higher usage limits and a SpaceX compute deal, Anthropic, May 6, 2026
• Privacy Policy, Anthropic, read Nov 6, 2026