Anthropic files go public, setting the stage for a huge IPO
Anthropic, the AI company behind the chatbot Claude, kept confidential on Monday for an initial public offering in what could be a once-in-a-generation moneymaking moment on Wall Street.
With its IPO filing, Anthropic is expected to be among three major companies set to go public this year, along with rocket companies SpaceX and OpenAI, which kicked off the AI boom in 2022 with its ChatGPT chatbot.
Their IPOs, which would be among the largest ever, could create a tsunami of investment and employee wealth and launch the world’s first billionaire in Elon Musk, who owns about 50 percent of SpaceX. Public offerings could also flood the nonprofit world with new money, as Anthropic and OpenAI have pledged much of their stock to charity. SpaceX’s IPO is expected this month, and OpenAI is preparing to file in the coming weeks.
The IPO filing also increases Anthropic’s competition with OpenAI. Last week, Anthropic officially passed OpenAI as the world’s top-flying AI startup with $65 billion in new funding, valuing it at $900 billion before including new capital. OpenAI’s last valuation was $730 billion.
In a statement, Anthropic said the filing “allows us to go public” after the Securities and Exchange Commission reviewed its filings. The company did not provide details on the timing or size of the IPO, saying the deal would “depend on market conditions and other factors.” With Monday’s filing, a public offering could happen as early as this fall.
Among the three companies seeking to go public, Anthropic is the youngest and fastest growing, thanks in large part to its AI tools for automatically writing computer code. Last week, Anthropic said its revenue run rate, which is its expected revenue for the year based on its current performance, topped $47 billion in May. It is unclear whether the company is profitable.
“The focus on coding is what makes this file interesting,” said Shashi Bellamkonda, director of Info-Tech Research Group, a technology research and consulting firm. “Anthropic didn’t try to be everything. No browser, no image generation, no business layer. This discipline is now $47 billion.”
Founded in 2021, Anthropic was founded in San Francisco by Dario Amodei, its CEO, and a handful of other researchers who worked with him on OpenAI. Since the company’s inception, Mr. Amodei has repeatedly described what he believes is the transformative nature of so-called artificial general intelligence, or AGI, a yet-to-be-achieved artificial intelligence that rivals the capabilities of the human brain.
Antropic also emphasized that this technology could be dangerous to humanity. This focus on security helped Anthropic maintain a positive public image, but it also led to tensions. This year, the Pentagon banned Anthropic’s technology from its systems because the company wanted to set some limits on what military and intelligence officials could do with it.
Recently, Anthropic’s relationship with the federal government has been in flux after it released a powerful model called Mythos that can identify and patch security vulnerabilities in software systems. Mr. Amodei visited the White House as intelligence agencies grappled with how Mythos was operating.
Other AI labs, such as OpenAI and Google’s DeepMind, have created a wider range of products, including AI-focused web browsers, business tools, and image-generating applications. But Anthropic has focused primarily on building AI that excels in coding software and tools for enterprise customers.
Last fall, that focus seemed to be paying off. Anthropic created a breakthrough model with Claude Opus 4.5, which the company described as “a meaningful step forward in what AI systems can do,” with an emphasis on software coding. Its usage skyrocketed and Anthropic’s revenue skyrocketed as more business customers started paying for the service.
“Claude’s latest advancements have led to widespread adoption among the world’s most demanding organizations,” said Brad Gerstner, founder of Altimeter Capital, an investor in Anthropic.
There has long been interest in owning a piece of Anthropic, with investment firms such as Menlo Ventures, Greenoaks Capital, Coatue and XN pouring money into it, as well as technology companies including Amazon, Google. Samsung, Micron and SK Hynix.
Much of the money Anthropic raised will likely pay for the rising cost of computing power, known as “compute,” a scarce resource given the growing demand for AI software. Mr Amodei said Anthropic sees “unprecedented demand” for its products, requiring it to pay for even more computing resources from companies like Microsoft, Amazon and Google to keep up.
Anthropic also faces stiff competition. OpenAI has refocused to focus on coding products using Codex, its software coding tool. Mr Musk, who runs SpaceX, has agreed to merge with and potentially acquire Cursor, a rival start-up focused on coding software. Larger companies, including Google and Microsoft, offer developers their own coding tools.
But SpaceX also became an Anthropic partner. Anthropic recently said it reached an agreement to use all of the computing power at the rocket company’s Colossus 1 data center in Memphis. The move gives Anthropic access to more than 220,000 AI chips and opens the door to a collaboration with SpaceX to create AI data centers in space.
(The New York Times sued OpenAI and Microsoft in 2023, alleging copyright infringement on news content related to AI systems. Both companies denied the claims.)
Cade Metz contributed reporting.