OpenAI IPO may be delayed until 2027? CEO Altman: No go unless valuation reaches $1 trillion.

OpenAI is reportedly considering delaying its IPO until 2027, primarily due to CEO Sam Altman's demand for a $1 trillion valuation, heavy energy costs, and increased scrutiny from the White House on new models.

According to the New York Times, OpenAI is leaning toward postponing its initial public offering (IPO) to 2027. External speculation points to recent volatility in tech stocks, the risk of significant price swings after SpaceX's record-breaking listing, and founder and CEO Sam Altman's insistence on achieving an unprecedented $1 trillion valuation. After OpenAI announced its intention to file this year, the actual public offering may be pushed back another year.

Altman Demands $1 Trillion Valuation; SpaceX's Break-Even Drop May Dampen Investor Enthusiasm

According to reports, Altman has indicated that no compromise solution below a $1 trillion valuation will be considered. However, internal voices at the company are also advocating for a delay to 2027, citing concerns over the company's heavy cash burn rate, an estimated $600 billion in basic computing infrastructure costs by 2030, and the company's lack of full preparedness for the disclosure obligations of a publicly traded company.

In its last private fundraising round, OpenAI was valued at around $850 billion. To successfully challenge the $1 trillion threshold in a public offering, the company would need not only strong revenue growth to bridge the valuation gap but also market enthusiasm for the IPO. However, the recent high-profile drop of SpaceX's stock below its offering price may make investors skeptical of giant IPOs in the short term.

White House Tightens AI Model Regulation; GPT-5.6 to Be Released via Customer-by-Customer Approval

On June 25, the White House demanded that OpenAI limit the rollout of its next-generation GPT-5.6 model to a small number of government-approved partners before a full-scale launch. Previously, the US Commerce Department restricted access to Anthropic's Mythos 5 and Fable 5, forcing Anthropic to fully disable both models.

However, the requirement for OpenAI differs from Anthropic's foreign access ban. Instead, GPT-5.6 must first enter a preview phase, with the government individually vetting enterprise customers. US Commerce Secretary Howard Lutnick directly discussed GPT-5.6 with Altman on Wednesday, hoping to ensure that all relevant government agencies have tested and approved the model.

Anthropic May Go Public First, Potentially Capturing AI Industry Pricing Power

Anthropic confidentially filed an S-1 registration with the SEC on June 1, aiming to list on the Nasdaq in October 2026 with a target valuation of $965 billion, with Goldman Sachs, JPMorgan Chase, and Morgan Stanley acting as lead underwriters. If OpenAI does indeed delay its timeline to 2027 and Anthropic proceeds as planned with its October 2026 listing, the underwriting banks' narrative that "whoever goes public first defines the industry" would position Anthropic as the market's focal point.

Anthropic expects to reach breakeven by 2028. In comparison, OpenAI is projected to face losses of up to $74 billion in 2028. However, due to factors such as Fed hints of interest rate hikes in the second half of the year, a shortage of AI memory, and the pullback in SpaceX's stock price, the market environment for Anthropic is expected to be more challenging than when it briefed investors in January.

• This article is republished with permission from: "Chain News"
• Original title: "OpenAI IPO May Be Delayed to 2027! CEO Altman: No Deal Below $1 Trillion Valuation"
• Original author: Co2