Anthropic, the maker of Claude, is pressing ahead with a blockbuster US stock market listing that could value the artificial intelligence company at about USD 1.3 trillion, even as it warns investors that advanced AI could pose “catastrophic or existential risks to humanity.”
The San Francisco-based company has set an initial public offering price range of USD 70 to USD 78 a share, offering 825 million shares and seeking to raise as much as USD 64.4 billion.
At the top of the range, its valuation would reach about USD 1.4 trillion on a fully diluted basis, including stock options and restricted shares, while its value excluding those securities would be about USD 1 trillion, according to its latest filing with the US Securities and Exchange Commission.
Trading is expected to begin on the Nasdaq on November 19 under the ticker ANT. Anthropic is also arranging a separate USD 26 billion private placement with strategic investors, taking the total proceeds from the IPO and related investment to as much as USD 90.4 billion, Reuters reported.
The prospectus carries an unusually stark warning from a company whose business depends on selling AI products. Anthropic says sufficiently advanced systems, if misused, replicated, weaponized, or allowed to operate with significant autonomy without safeguards, could cause catastrophic or existential harm.
It flags risks, including cyber attacks, biological misuse, harmful autonomous actions, and models behaving in ways their developers cannot reliably predict or control.
In its latest amended filing, Anthropic disclosed that it had crossed an internal threshold under its Responsible Scaling Policy for AI systems accelerating research and development.
The company said the determination was based on what it expects from future models and its overall pace of progress, rather than capabilities it has already observed in current models.
Anthropic said AI tools were already speeding up its research, and that further acceleration could outpace its ability to develop and implement safety, alignment, and security measures. It is assessing the implications and said it would publish further reports.
The threshold does not mean its systems are independently building successor models; it signals concern that AI could increasingly contribute to developing more capable AI.
The company cautions that a serious incident could damage its reputation, trigger government intervention, and undermine customer confidence. It also says no single insurer, and potentially the insurance industry as a whole, may have enough capacity to cover the losses from a catastrophic AI-related event.
The financial figures underline both the opportunity and the scale of the bet. Anthropic reported run-rate revenue of about USD 81 billion as of October 31, up USD 10 billion in a month, according to the filing. It had more than 11,900 customers generating at least USD 100,000 in annualised revenue as of September 30, compared with more than 950 a year earlier.
However, its accounts show a net loss of USD 104 billion for the nine months ended September 30, 2026. That figure includes a USD 94.3 billion non-cash charge linked mainly to the increase in fair value of convertible notes as the company’s valuation rose.
The headline loss, therefore, does not represent an equivalent cash outflow from operating the business but highlights the complexity of its financing structure and the volatility that can affect reported results.
Anthropic’s growth also requires vast infrastructure investment. Reuters reported that the company has nearly USD 900 billion in computing commitments with partners including Amazon and Google, reflecting the long-term cost of securing the chips, data centers, and electricity needed to train and operate its models.
These commitments expose it to financing and supply-chain risks.
The IPO proceeds could help fund that expansion and reduce reliance on private capital. Google, Nvidia, and AMD are among the cornerstone investors in the separate placement, while Broadcom, Micron, Microsoft, and Taiwan Semiconductor are also participating in the wider investment arrangements, Reuters reported.
Founded in 2021 by chief executive Dario Amodei, his sister Daniela and former OpenAI employees, Anthropic has positioned itself as a safety-focused rival. Claude has gained traction among businesses for coding, analysis, and workplace applications.
The listing will also be watched for what it says about rival OpenAI, which is preparing for a potential public offering of its own.
A strong debut could validate the enormous valuations attached to frontier AI developers; a weak reception could force investors to reassess the sector’s capital needs and profit prospects.
Anthropic is incorporated as a public benefit corporation, a structure requiring it to balance shareholders’ financial interests with its purpose of developing advanced AI responsibly for humanity’s benefit.
The arrangement will be watched as public investors seek influence over a company with consequential technology.
For now, Anthropic is asking public-market investors to back both its commercial promise and its warnings about the technology it is building. Whether investors regard those disclosures as evidence of responsible transparency or a reason to demand a lower valuation will be tested when the shares begin trading.