Recently, the leading figure in the field of artificial intelligence, OpenAI, is engaging in new private financing discussions with investors, aiming for a valuation of $1.2 trillion. This proposed valuation represents a significant increase from the $85.2 billion post-money valuation established in March this year following a $12.2 billion funding round, and will continue to keep the company's fundraising scale at the forefront among private tech companies. However, this high valuation depends on its initial public offering (IPO) reaching a similar level by 2027. Originally, the advisory team had considered an earlier listing in the third quarter with a lower valuation, but after comprehensive assessment, the company prefers to delay the issuance until 2027, with CEO Sam Altman clearly stating that $1 trillion is the minimum threshold for the IPO.

In terms of capital, existing investors are deeply involved. As the largest single institutional shareholder, Microsoft has invested over $13 billion to hold about 27% of the shares; NVIDIA has invested around $30 billion in the form of computing capacity. Additionally, SoftBank anchored its investment with a $4 billion bridge loan in March, while Amazon's $5 billion anchoring investment is closely related to the IPO or milestones in general artificial intelligence. It should be noted that the $12.2 billion surface number includes a large number of conditions, deferred or supplier-related rights, such as NVIDIA's $30 billion primarily being computing capacity used to offset GPU infrastructure costs rather than pure cash.

From a financial perspective, OpenAI's revenue shows strong growth momentum. By February 2026, its annualized revenue reached approximately $25 billion, growing by about 92% compared to the past 12 months, with first-quarter revenue at $5.7 billion, and the full year moving towards a $30 billion target, with some estimates suggesting it has already approached $40 billion by mid-2026. Among these, the enterprise business has become the core growth engine, accounting for more than 40% of revenue, and is expected to match consumer business by the end of the year.

However, along with the high growth comes an expanding loss. The operating loss in the first quarter of 2026 was approximately $9.3 billion, increasing to about $12.3 billion in the second quarter, with the full-year operating loss expected between $27 billion and $33 billion. With the cost of reasoning rising to $14.1 billion in 2026, plus the company's strategic layout to achieve 30 gigawatts of computing capacity by 2030, computing and talent expenses grow in tandem with revenue. Although the gross profit margin has risen from about 33% a year ago to about 39% in the first quarter of 2026, every efficiency improvement brings profit space which is quickly reinvested into the next round of production and research, causing the operating loss to continue to worsen. At the current revenue and $1.2 trillion valuation, its price-to-sales ratio is around 40 times, and the massive losses without a credible path to profitability have triggered market scrutiny of its high multiple pricing.

In terms of user base and market competition, OpenAI still holds a dominant position. By February 2026, ChatGPT had 900 million weekly active users and exceeded 1 billion monthly active users in May, with over 50 million individual subscribers and 9 million paid enterprise users, with 92% of Fortune 500 companies using the service. However, its market share faces serious challenges, with Sensor Tower data showing that its share in global AI assistants fell below 50% in May to about 46%. At the same time, competitor Anthropic has surpassed in enterprise-level API spending, with its annualized revenue reaching $30 billion in April 2026 and surpassing OpenAI. Intense market competition has prompted OpenAI to revise its product roadmap twice within half a year and adjust the release plans of certain features.

Regarding the nature of the losses, the core market divergence lies in whether they are cyclical expansion costs or structural characteristics that cannot be self-corrected. Analysis shows that OpenAI's losses exhibit strong structural characteristics: on one hand, the cost of reasoning grows rigidly with the scale of usage; on the other hand, the company has locked in a fixed cost base through a 30-gigawatt commitment that must be fulfilled regardless of demand. This model, driven by continuous computing consumption and large infrastructure expenditures, means high losses are not a short-term phenomenon.

Supporters argue that each technological platform transition has given rise to dominant giants whose value far exceeds expectations, and ChatGPT's large user base and enterprise penetration are indicative of infrastructure-level adoption. On the other hand, opposing views worry that, in the context of model capabilities converging and intensifying competition, the high cost base may put it at risk of a price war. Overall, the $1.2 trillion valuation is not only a bet on OpenAI's own revenue growth, but also a belief pricing on the continued expansion of the entire AI capital expenditure cycle. The key in the future lies in whether the gross profit margin can further break through, whether enterprise revenue can continue to catch up, and whether the loss trajectory can see substantial narrowing.