Anthropic Submits S-1 Filing to SEC, Officially Launches IPO

Article body

On October 26, Anthropic, the artificial intelligence company that develops the Claude series of large models, submitted an S-1 registration statement to the U.S. Securities and Exchange Commission (SEC), formally initiating the initial public offering (IPO) process.

The company plans to list on the Nasdaq Global Select Market under the ticker symbol ANT. The number of shares to be offered, the offering price range and the listing date have not yet been determined, and the offering is being underwritten with Goldman Sachs and Morgan Stanley as lead underwriters.

Screenshot of the prospectus (Source: SEC website)

The prospectus shows that Anthropic was founded in 2021 by siblings Dario Amodei and Daniela Amodei, along with several former core members of OpenAI. The company focuses on the Claude series of large models, enterprise API services and the Claude Code programming tool, providing generative AI capabilities to B2B clients in finance, technology, government and enterprise sectors. Amazon, Google, JPMorgan Chase and Salesforce are all among its core customers, and it currently has more than a thousand enterprise customers with annual spending exceeding $1 million.

Anthropic stated that the proceeds from this IPO will be used mainly for frontier large model research and development, computing infrastructure construction, AI safety research, product iteration, global market expansion and supplementing working capital. The market expects this IPO to potentially become the largest technology IPO in the world.

Explosive revenue growth, with computing investment bringing huge losses

The financial data is the part of the prospectus that has drawn the most market attention. The "condensed consolidated operating data" table in the prospectus shows that Anthropic's commercialization began in 2023, when full-year revenue was only $41 million; it jumped to $386 million in 2024 and further reached $4.585 billion in 2025. Entering 2026, growth accelerated again, with first-half revenue already reaching $16.286 billion, more than 3.5 times the full-year 2025 figure.

In contrast to the rapid revenue growth is the company's continued huge losses. The prospectus shows that under GAAP, the net loss for the first half of 2026 was $80.748 billion. The company explained that this huge loss mainly came from "changes in the fair value of convertible notes, derivative liabilities and SAFEs," which amounted to $72.388 billion; the reason is that the company's valuation continued to rise during each reporting period, causing the fair value of liability instruments such as preferred shares and convertible bonds to expand accordingly. The company stressed that this is a non-cash accounting loss and does not correspond to actual cash outflow. For the full year 2025, the loss from the same item was $34.308 billion, and the full-year net loss was $41.965 billion.

If non-cash items such as share-based compensation are excluded, under the company's self-defined non-GAAP measure, adjusted operating profit for the first half of 2026 was positive at $1.618 billion, with an adjusted operating margin of about 10%, the first half-year profit to appear in the prospectus.

As of June 30, the company had $112.4 billion in cash and equivalents and $19.3 billion in short-term investments, totaling $131.7 billion.

Special voting rights structure

As a Public Benefit Corporation (PBC), Anthropic retains its public-benefit character and, beyond commercial goals, continues to incorporate AI safety and the public interest into its corporate governance objectives.

In terms of equity voting design, the company established a special governance entity, Founder LLC, composed of seven co-founders. The Class F shares held by this entity carry 50.1% of the company's voting power and can dominate key matters such as director elections and major strategy; the Class A common shares issued to public investors carry only one vote per share, and external investors cannot control the company's core direction. This structure is intended to ensure that the AI safety strategy and long-term technology roadmap are not disrupted by short-term capital market pressure, and it also means that ordinary investors' voting rights are significantly limited; the prospectus lists this as an important risk factor.

Unlike past technology company IPO materials, Anthropic devotes a large amount of space in its prospectus to warning about the potential risks of frontier large models, and it is also the first time an AI company has systematically mentioned existential AI risks in a formal listing document aimed at public investors. The document warns that more capable frontier AI systems may develop self-preservation tendencies, with risks including refusing shutdown, concealing information and manipulating outputs, and in extreme cases catastrophic risk. The company said that alignment, safety evaluations and red-team testing will continue to be embedded throughout the model development process, and that the safety team has veto power over the launch of high-risk models; at the same time, the company commits that part of its equity returns will be used for AI safety public-benefit projects, fulfilling the commitments of its public benefit corporation charter.

Multiple investment institutions take stakes

The prospectus discloses that, as of June 30, the company had raised a cumulative $141.1 billion.

Strategic investors include Amazon, Alphabet, Salesforce, Microsoft and Nvidia. Among them, Amazon has actually contributed $18 billion, with an additional conditional contribution quota of up to $15 billion (triggered by computing milestones); Alphabet has cumulatively committed up to $40 billion, with $10 billion in cash already in place and $30 billion tied to model performance indicators, but the equity cap is contractually locked at 15%; Microsoft and Nvidia have respectively committed up to $5 billion and $10 billion in investment, with part of the funds already implemented in the G round, but the prospectus did not disclose the corresponding shareholding ratios.

In addition, the company's shareholder list also includes a number of internationally renowned investment institutions such as Fidelity, Lightspeed, BlackRock, Blackstone, TPG and Baillie Gifford, while sovereign funds including GIC, Qatar's QIA, Abu Dhabi's MGX, Ontario Teachers' Pension Plan and Temasek have also participated in the investment.

(Source: Securities Times) Mid-article embedded advertisement