OpenAI has held early talks with major investors about a funding round that could value the ChatGPT maker at roughly $1.2 trillion ahead of an initial public offering, Reuters reported on September 15, citing the Financial Times. The proposed figure would place the company about 41% above the $852 billion post-money valuation set in its most recent funding round in March.

The discussions point to a larger question facing the sector: how much value frontier AI companies generate before public investors can access them. As compute costs increase and markets continue to consolidate, the central issue is whether enterprise-level demand and a historic boom in AI investment can keep pace with trillion-dollar private-market valuations.

Investors, not OpenAI, initiated the talks

The negotiations remain at an initial stage, and the valuation could still change. According to Reuters, investors approached OpenAI rather than the company starting the process itself. OpenAI did not respond when approached for comment.

The news also follows a public signal from the company's leadership. On September 12, CEO Sam Altman said that OpenAI would not pursue an IPO in 2026, citing concerns over frontier AI safety. Both OpenAI and Anthropic have endorsed stronger regulations on more powerful AI systems. Read together, the two developments frame the stakes of the current talks: with management signaling no IPO in 2026, a completed private round would determine who holds exposure to OpenAI ahead of any eventual public listing.

From $852 billion to $1.2 trillion in six months

According to OpenAI, its most recent funding round, held in March, netted $122 billion in committed capital and produced a post-money valuation of $852 billion. Amazon, NVIDIA and SoftBank provided funding for the round, while partner Microsoft continued to participate. An additional $3 billion was raised from other investors through bank routes.

At the time, OpenAI said it was earning $2 billion in revenue each month, with enterprise customers generating more than 40% of that revenue and the revenue forecast to reach similar levels by the end of the year.

Measured against that baseline, the proposed $1.2 trillion valuation represents a 41% increase in less than six months.

The private mega-cap club OpenAI is racing inside

According to Forge Global, Anthropic, OpenAI and xAI have each reached market values above $100 billion in a far shorter period than earlier generations of private companies. xAI took only 2.3 years to reach that milestone and Anthropic about 4.5 years, compared with roughly 16 years for companies established before 2011.

Meanwhile, DefiLlama's pre-IPO tracker estimated, as of September 14, a valuation of $1.48 trillion for Anthropic and $903.29 billion for OpenAI. These figures are estimates of pre-IPO value rather than the valuations set in either company's latest funding round. On that basis, Anthropic sits approximately 64% above OpenAI. If OpenAI were valued at the proposed $1.2 trillion while Anthropic's estimate held steady, the gap would narrow to nearly 23% — Cryptopolitan's calculations show how quickly another mega funding round could reshape private-market predictions.

Cryptopolitan previously reported that Anthropic's annualized revenue run rate has surpassed OpenAI's, indicating that investors are increasingly factoring enterprise growth and future market share into valuations, alongside model capability.

A $1 trillion capital wave beneath the valuations

Goldman Sachs Research forecasts that AI investment worldwide will exceed $1 trillion in 2026, with $581 billion of that total going to the United States alone. Gartner expects global expenditure on AI platforms and models to reach $64.25 billion, a 63.4% increase compared with 2025.

"Enterprise AI budgets are coming under greater scrutiny, with increased focus on usage efficiency, cost control and measurable outcomes." — Arunasree Cheparthi, Gartner

That tension is significant. Spending on infrastructure projects may continue to rise even as customers demand clearer returns. A survey carried out by McKinsey shows that 40% of respondents from large companies actively implementing AI agents, compared with only 22% at small companies. The OECD has also said that hardware and data are largely concentrated in a few hands, while research by S&P Global suggests that economies central to AI development have exceeded the global benchmark rate every month since the end of 2025.

Why the IPO may be a liquidity event, not a launchpad

Forge argues that for companies already worth more than $100 billion, much of the value creation now happens while they remain private. That makes an IPO look increasingly like a liquidity event for existing investors rather than the beginning of a new phase of major valuation growth.

Anthropic may test that idea first. Reuters says the company could begin marketing an IPO in mid-October at the earliest and complete the listing shortly before the U.S. midterm elections in November. If so, OpenAI's $1.2 trillion talks are less about what an IPO might unlock than about how much value private investors are willing to price in before one happens. The near-term markers are concrete: whether Anthropic's listing follows the timetable Reuters outlines, and whether OpenAI's discussions harden into a signed round at or near the proposed figure. Either outcome would set the next reference points for frontier AI valuations as the sector's largest private companies move closer to public markets.