Artificial intelligence giant Anthropic files papers to sell shares to the public for the first time

SAN FRANCISCO (AP) — Anthropic, a major artificial intelligence company, filed paperwork Monday to sell shares to the public for the first time. Its stock market debut could be the biggest ever, even as its top executive warns that advanced AI could threaten humanity's existence.

The filing puts the San Francisco company on track to enter the stock market next month. Its public offering could be bigger than the SpaceX offering in June, which raised a record $86 billion. Anthropic could also go public before OpenAI, setting up what are expected to be two huge stock market debuts by rival AI companies.

The filing comes weeks after CEO and co-founder Dario Amodei argued in an essay that slowing the development of the most advanced AI models could give Anthropic huge resources to speed up its growth. The amount the company will ultimately raise is still uncertain, as are its final value and share price. But even at lower estimates, the five-year-old startup could become one of the world's most valuable companies. Its public offering could also increase the wealth of Amodei, his sister and co-founder, Daniela, by billions of dollars.

The 223-page document showed that Anthropic's revenue is growing quickly. But, like many other technology companies preparing to go public, it has not yet made a profit. Anthropic, which makes the Claude chatbot, reported a net loss of $5.6 billion in the first half of 2026. That compares with a loss of $2.7 billion during the same period a year earlier.

Despite its losses, Anthropic, which private investors valued at $183 billion earlier this year, is reportedly aiming for a value of $1.5 trillion or possibly $2 trillion, although its final value has not been set. At $1.5 trillion, Anthropic would rank among the world's most valuable publicly traded companies, alongside Tesla.

Reports say the stock offering could raise as much as $100 billion, beating the current record set earlier this year by another one of Elon Musk's companies, SpaceX.

In a regulatory filing, Anthropic said it would list its shares on the Nasdaq Global Select Market under the stock symbol "ANTH."

Risks are plentiful, but the founders could get a big payoff

In its filing, the company said it could use some of the money it raises to buy other businesses and technology, as well as invest in the equipment and systems it needs to grow and compete with rivals.

Founded in 2021 by former OpenAI leaders, Anthropic is locked in a battle for dominance with OpenAI and its two leaders, CEO Sam Altman and Amodei, who are widely seen as fierce rivals.

They also face many other competitors, including Google, newer startups in the U.S. and China, and Musk's SpaceX, which has an AI business. Anthropic, OpenAI and SpaceX's AI division have all been losing more money than they earn because of the huge costs of building AI. This has raised concerns about a possible bubble.

The IPO would help cover any cash shortfall and could bring extraordinary wealth to Anthropic's seven founders, including Amodei and his sister. Together, they own nearly 500 million shares, though the offering document does not make clear how much of the company they will own after the IPO.

The filing also listed many risks for the San Francisco-based company. These include the possibility that it may not be able to keep key employees, deal with regulatory barriers or secure enough computing power to run its AI systems.

But the biggest warning was that its AI models could "pose catastrophic or existential risks to humanity."

Calls for a slowdown in AI development

The IPO comes as worries grow both inside and outside the AI industry. Reports continue to emerge about increasingly powerful systems that can solve problems beyond human ability but could also act on their own and carry out more harmful tasks.

In September, Amodei said the AI industry should slow its fast-moving development to give safety measures time to catch up. Without such a slowdown, Amodei warned that AI could be capable of letting loose a swarm of agents that could take over the entire internet within six to 12 months.

He offered a plan in a post on his website to increase checks on the industry, noting that Anthropic is already taking some of these steps on its own. But he said other actions would require cooperation across the industry and with governments around the world, including authoritarian governments.

"I believe that if slowing down bought us even an extra year or two before models reach critical levels of capability, and we used that time to advance alignment, we could greatly reduce the risk that something goes seriously wrong," Amodei wrote.

Copyright 2026 The Associated Press. All rights reserved. This material may not be published, broadcast, rewritten or redistributed without permission.