OpenAI is reported to have opened preliminary talks to raise up to $100 billion at a valuation of around $750 billion, a move that would represent a sharp escalation in the scale of capital flowing into the artificial intelligence sector and place the company at more than double the valuation of its closest rivals.

According to reports, the discussions signal a roughly 50% jump from the $500 billion valuation confirmed in October, when current and former employees sold around $6.6 billion worth of shares in a secondary transaction. If realised, the figure would rank among the largest private company valuations ever discussed and underscore investor appetite for exposure to frontier AI development.

The proposed fundraising is closely tied to OpenAI’s capital-intensive infrastructure ambitions, most notably the $500 billion “Stargate” project backed by SoftBank and Oracle. The initiative is designed to secure the physical computing layer required for advanced AI and artificial general intelligence development, as the company faces rising competitive and financial pressures.

Reports suggest the Stargate strategy relies on a “circular revenue” model, under which OpenAI raises capital to purchase large volumes of hardware, including Nvidia GPUs and AWS Trainium chips. That compute capacity is then either consumed internally or rented back to the market, effectively recycling capital through infrastructure deployment and usage.

At a $750 billion valuation, OpenAI would be worth around 2.5 times the size of Anthropic, which is currently pursuing its own discussions with bankers around a potential IPO. SoftBank’s role remains a key variable, with reports indicating a possible $19 billion commitment to the broader Stargate vehicle, reinforcing Masayoshi Son’s position as a central financier in the generative AI boom.

Sarah Friar, OpenAI’s chief financial officer, has framed the scale of the investment challenge in national rather than purely corporate terms. “American strength in technology will come from building real industrial capacity which requires the private sector and government playing their part,” she said.

The fundraising discussions come against the backdrop of intensifying competition in consumer and enterprise AI. OpenAI chief executive Sam Altman has reportedly declared a company-wide “Code Red” following the rapid growth of Google’s Gemini AI, which has reached 650 million monthly active users. The move marks a shift away from broad feature expansion towards a narrower focus on model quality, stability and efficiency.

Fidji Simo, CEO of Applications at OpenAI, described the change in emphasis as a matter of prioritisation rather than retrenchment. “We announced this code red to really signal to the company that we want to martial resources in one particular area, and that’s a way to really define priorities and define things that can be deprioritized,” she said.

As part of this efficiency drive, OpenAI has ended the “feature drop” approach that characterised earlier growth phases and introduced changes affecting free-tier users. These users have been defaulted to the cheaper GPT-5.2 “Instant” model, with automatic routing removed as the company seeks to reduce inference costs.

Financial disclosures underline the urgency of these measures. Internal projections point to a net loss of $14 billion in fiscal year 2026, driven by escalating training and operational expenses. With profitability now expected to be pushed back to 2029, OpenAI is seeking to bridge the gap through large-scale external funding and more traditional revenue strategies.

To that end, the company has appointed former Slack CEO Denise Dresser as chief revenue officer. Her remit includes building an enterprise-focused sales operation capable of securing high-value, multi-year contracts to help offset the projected deficit.

At the same time, OpenAI is attempting to diversify its technical dependencies as part of an ongoing arms race in AI infrastructure. Talks with Amazon around a potential $10 billion investment are reported to include commitments to use AWS Trainium chips, marking a strategic hedge against reliance on Nvidia hardware. Dave Brown, vice president of compute at AWS, has emphasised the transactional nature of the arrangement.

On the product side, the recent launch of GPT-5.2, comprising Instant, Thinking and Pro variants, represents a direct response to Google’s Gemini 3. The models are positioned with pricing that undercuts key competitors, alongside developer tooling updates designed to mirror open standards and retain platform loyalty.

Altman has characterised rapid reaction to rivals as defining the current phase of competition. “I believe that when a competitive threat happens, you want to focus on it, deal with it quickly,” he said.

Recommended reading

• UK Launches “World’s First” AI Safety Institute
• The AI Vendor Race is Revolutionising the Tech Landscape
• UK AI Security Institute Launches New £15M Project
• AI Key for National Security, Says New Turing Institute Report

Beyond private fundraising, OpenAI is also laying the groundwork for a potential public listing. Reuters has previously reported that the Microsoft-backed company is preparing for a possible IPO that could value it at up to $1 trillion, with a filing potentially as early as the second half of 2026. Such a move would place OpenAI among the most valuable companies in the world and reflect investor expectations that AI will become a foundational technology across sectors.

The reported valuation discussions sit within a broader surge in AI dealmaking. Anthropic’s valuation has climbed to an estimated $350 billion following recent investments, while other players such as Perplexity and xAI have also attracted multi-billion-dollar valuations. According to CB Insights, more than 1,300 AI startups now carry valuations above $100 million, including nearly 500 valued at $1 billion or more.

Despite continued enthusiasm, investors are closely watching whether demand for AI services can sustain the extraordinary levels of capital spending now required. Global AI spending is expected to reach $375 billion this year and rise to $500 billion in 2026, reinforcing both the scale of opportunity and the risks associated with an increasingly capital-heavy race for technological leadership.