Following Anthropic, OpenAI is secretly filing for an IPO

OpenAI, the maker of ChatGPT, has filed confidentially for an initial public offering, the company announced Monday in a blog post. The filing comes just over a week after its main competitor, Anthropic, filed to go public, intensifying the race between the two AI companies.

OpenAI, recently valued at $852 billion, has filed a draft registration statement with the U.S. Securities and Exchange Commission for a proposed initial public offering. The company did not mention the number of shares or specify a price.

The filing is the latest sign that 2026 will be a successful year for the public markets. SpaceX is also expected to debut at a valuation of $1.75 trillion, meaning three of the most closely watched companies in technology could go public within months of each other — a concentration of high-risk offerings the markets haven’t seen in a generation.

OpenAI is racing to go public even though it recently failed to meet its own goals for new users and revenue, according to the Wall Street Journal. Its CFO, Sarah Friar, has reportedly raised concerns that OpenAI may not be able to support its massive data center spending. The burning appears to be massive.

In late March, OpenAI secured $122 billion in the largest funding round in Silicon Valley history — $3 billion of which came directly from retail investors via banking channels. But the company expects to spend about the same amount on computing power for AI research alone in 2028, and the projects are burning $85 billion that year even after doubling sales from the previous year, according to the Wall Street Journal. For context: OpenAI is asking public market investors to buy into a company that, according to its own projections, won’t generate more money than it spends for at least another four years.

SpaceX offers a parallel data point. Its spending on AI, while not enormous, shows how the cost of training large language models can exceed the revenues generated by those models — a structural challenge that the entire industry is grappling with, and one that public market investors will have to price in.

On the other hand, Anthropic presented investors with a rosier picture of its financials, saying it was close to achieving its first quarterly profit. However, with a recent $65 billion funding round and another $36 billion in chip debt on the way, Anthropic’s burn rate isn’t exactly modest.

A confidential IPO filing allows OpenAI to begin preparing it for a public offering without publicly revealing detailed financial information or business risks, which is why the company hasn’t shared share prices or how much it hopes to raise yet. However, secondary markets provide a glimpse into what investors are willing to pay.

Anthropic’s valuation recently soared to $1 trillion on Forge Global, a retail secondary market platform, surpassing OpenAI, which was clocked at about $880 billion in April.

Anthropic’s appreciation rate is far outpacing OpenAI this year — 123% year-to-date versus 11.3% for OpenAI, said David Shapiro, founder and CEO of OpenVC and overseer of the NYSE OpenVC 500 Index, which tracks the largest public and private companies in the United States. However, despite Anthropic’s clear boost, OpenAI is seeing no shortage of secondary interest.

“From a secondary investor perspective, OpenAI has really grown into a big part of its valuation,” Shapiro told TechCrunch. “We haven’t seen an OpenAI pit or anything close, and the evaluation remains very successful, according to the indicator.”

He added that OpenAI’s shares on the secondary market “have seen a slight rally over the past few days, suggesting that investors may be pricing both as ‘double winners’ in the broader LLM race.”

But the race to get to public markets first is a real concern. Whoever debuts first, experts say, will likely seize more of the capital that has become increasingly scarce for AI companies — much of which SpaceX may have already absorbed, and which is expected to go public among the three companies.

Additionally, Anthropic’s filing disclosures will identify valuation firms restricting how OpenAI prices its own offerings when it files, according to a recent PitchBook report that described OpenAI as overvalued relative to its fundamentals. In other words, if Anthropologie prices conservative, OpenAI’s path to its target valuation becomes more difficult.

OpenAI, founded in 2015 as a non-profit research lab, revolutionized the world of AI when it released ChatGPT in 2022, sparking a wave of major advances in language models across the industry.

While OpenAI has expanded its products to accommodate enterprise and government customers, the company has a strong reputation for being more consumer-focused than its competitor Anthropic. The company has built real scale, with around 900 million weekly active users.

The IPO comes after major internal conflicts within the company. In 2022, OpenAI’s board of directors fired Altman over what it described as a lack of transparency and concerns about whether he was committed to the company’s mission of benefiting all of humanity. Altman was quickly reinstated, and board members who participated in the coup, including co-founder Ilya Sutskever, left shortly after. This incident raised governance questions that have never been fully resolved and will likely be closely scrutinized by potential public investors.

Recently, OpenAI has faced several lawsuits, including a recent lawsuit from Florida that accuses the company and Altman of harming children by providing information to school shooters, providing guidance on self-harm, and promoting addiction among young users. The Florida complaint adds to a series of lawsuits filed against OpenAI and other chatbot makers in the wake of user delusions, self-harm, suicide and mass casualty events.

Last month, OpenAI won court after co-founder and competitor Elon Musk sued the company and Altman over an alleged promise to keep the company a nonprofit. The case was ultimately dismissed after a jury and judge found that Musk had waited too long to file the suit, and had exceeded the statute of limitations when he filed the case in 2024.

OpenAI also faced criticism after its president, Greg Brockman, and his wife donated $12.5 million to Leadership Future, a pro-AI political action committee dedicated to thwarting local politicians who advocate for AI regulation. Both have also made similar contributions to MAGA Inc., a major pro-Trump political action committee. OpenAI has tried to distance itself from what it calls Brockman’s personal donations, saying the money was not given on behalf of the company.