OpenAI plans new funding round at potential US$1.2 trillion valuation

·by Henderson·AI
OpenAI plans new funding round at potential US$1.2 trillion valuation

According to foreign media reports, OpenAI is in preliminary discussions with major investors about a new private funding round targeting a valuation of approximately US$1.2 trillion, as it prepares for a future initial public offering (IPO). If the round is completed, it would represent an increase of roughly 40% over its post-money valuation of US$852 billion following its previous funding round in March. The direct driver of OpenAI's higher valuation is its recent strong revenue growth. People familiar with the matter said that, following the release of new models such as GPT-5.6 and Astra, the company's revenue growth has accelerated significantly, with annualized revenue exceeding US$40 billion — roughly double the roughly US$20 billion level recorded at the end of 2025.

OpenAI is raising a new private round to prepare for an IPO

In its previous funding round, OpenAI set a global record for AI company financings at a total size of US$122 billion, with a post-money valuation of US$852 billion, itself a record for the pace of valuation growth in the technology industry. People familiar with the matter said the negotiations remain at an early stage and that the final valuation and round size could change. Asian investors are also closely watching how major existing shareholders such as SoftBank respond to the new round.

Notably, this round is taking place against the backdrop of an OpenAI IPO delay. On September 12, OpenAI CEO Sam Altman told Fortune in an interview that the company would not go public in 2026. When asked whether 2027 was a more likely target, Altman said only that "let's just say it's not happening in 2026; there's a lot more work to do, including the safety and alignment work needed to deal with this critical moment." OpenAI confidentially submitted a draft IPO prospectus in June of this year.

Deutsche Bank Research has previously forecast that, if it lists at a valuation above US$1 trillion, OpenAI would rank as the 14th most valuable company in the world by market capitalization.

The decision to postpone the listing gives long-term investors such as SoftBank and Thrive Capital room to expand their holdings, but also defers the liquidity event that earlier backers have been anticipating. Another key factor driving OpenAI to accelerate its fundraising is the persistent pressure from competitor Anthropic. Anthropic's annualized revenue has surged from roughly US$1 billion in January 2025 to more than US$30 billion in April 2026, overtaking OpenAI on a revenue basis.

In terms of enterprise customer market share, 34.4% of surveyed enterprise customers in April 2026 were paying for Anthropic's AI products, surpassing OpenAI for the first time.

Anthropic is now preparing to launch its own IPO as early as October, targeting a valuation as high as US$2 trillion. The listing race between the two AI heavyweights is pushing the global AI capital competition to new heights. Despite strong revenue growth, the logic underpinning OpenAI's valuation still faces multiple tests. The company is projected to post a full-year loss of roughly US$14 billion in 2026, with cash burn set to widen further, and is not expected to achieve positive cash flow until 2030 at the earliest. Its gross margin is only about 33%, and inference costs are set to reach US$8.4 billion in 2025 and an estimated US$14.1 billion in 2026.

Some analysts note that, to provide a reasonable basis for investing in OpenAI's current round, one would have to assume an IPO valuation of more than US$1.2 trillion — an assumption that becomes increasingly untenable against the backdrop of Anthropic offering alternatives at lower prices. Deutsche Bank explicitly warned in its report that investors still have a very limited understanding of AI companies' long-term business models and unit economics, and that the IPO will serve as an "important litmus test" of whether the AI boom can deliver lasting economic value.

H
About the author
Henderson