The Wall Street Journal: OpenAI misses user and revenue targets, CFO worries about paying data center bills before IPO

📄Full Article· Automatically extracted by trafilaturaGemini 翻譯1909 words

The Wall Street Journal reported that OpenAI has failed to meet its internal goal of reaching 1 billion weekly active users for ChatGPT by the end of 2025, and has missed its monthly revenue targets multiple times in early 2026. The report notes that OpenAI CFO Sarah Friar has privately expressed concerns to executives that if revenue growth does not accelerate, the company may be unable to cover the massive data center contract obligations it has committed to. (Context: Sam Altman reportedly promised to invest $600 billion in computing power expansion to "accelerate OpenAI's year-end IPO," while the CFO worries about cash depletion within five years.) (Background: Is the partnership between Microsoft and OpenAI hitting a snag? The Wall Street Journal: Rifts are widening, and the "Golden AI Alliance" is drifting apart...) A company's valuation is essentially a collective bet on the future. The foundation of that bet is the belief that its growth is sustainable. But when growth figures begin to miss the mark, the foundation of that bet can start to loosen. The Wall Street Journal reported that OpenAI's internal goal for ChatGPT last year was to reach 1 billion weekly active users by the end of 2025, equivalent to one-eighth of the global population, but this milestone was not achieved. More recently, in early 2026, OpenAI also failed to meet its self-imposed monthly revenue targets for several consecutive months. WSJ reporters pointed out that the gap in both users and revenue for OpenAI is primarily due to a clear competitor: Anthropic. In two key battlegrounds—coding tools and the enterprise market—Anthropic is steadily expanding its share. According to figures tracked by multiple research institutions, Anthropic's enterprise API market share has grown from 12% in 2023 to approximately 32%; the gap in annualized revenue between the two companies has also narrowed from a wide margin to about $6 billion. The reason behind this is not complex: the Claude model family has reached or even surpassed the performance of GPT-4o in multiple metrics for coding tasks. Enterprise buyers are not concerned with which company has more name recognition, but rather whose model is more useful in their own workflows. OpenAI CFO Sarah Friar stated publicly at the beginning of the year that the priority for 2026 is "practical application." Put simply, the growth rate of ChatGPT consumer subscriptions is no longer a robust foundation for financial models; enterprise customers are. Enterprise customers accounted for about 40% of OpenAI's overall business at the beginning of this year, and Friar's goal is to push that to 50% by the end of the year. This goal itself speaks volumes: OpenAI has not yet reached the position it needs to be in before an IPO. The core tension in this Wall Street Journal report lies not just in the numbers, but in a cognitive gap between top executives. Friar has privately expressed concerns to other executives: if the company's revenue growth does not reach a certain threshold, OpenAI may not be able to pay for the data center contract obligations it has committed to. Over the past year, OpenAI has embarked on a massive expansion, making commitments to multiple data center projects totaling an estimated $600 billion—including installment payments over the next few years, rather than a one-time payment. The contractual commitments have been made, but the revenue growth supporting these commitments currently looks slower than planned. The board is also applying pressure. OpenAI's directors are scrutinizing these data center deals and questioning the decision-making logic of CEO Sam Altman, who continues to aggressively seek more computing power investment despite signs of softening revenue. This is a structural disagreement: Altman's logic is that more computing power is a prerequisite for next-generation models and a moat for competitiveness; Friar's logic is that contractual obligations will not automatically disappear just because models get stronger—cash flow is what matters. The financial figures are equally grim. According to estimates from multiple institutions, OpenAI is expected to lose about $14 billion in 2026, with an annual burn rate that could reach $57 billion in 2027, and it is not expected to break even until 2030. These numbers are not secrets, but against the backdrop of consecutive missed targets, they are being re-interpreted with greater scrutiny. OpenAI is racing toward a specific deadline: a public listing by the end of 2026. The logic of an IPO is that valuation requires a story, the story requires growth, and growth requires user and revenue numbers that continue to trend upward. When both of these numbers miss the mark, the foundation of the entire narrative begins to shake. The core of the problem is not whether OpenAI is still in the lead, but the margin of its lead and the speed of capital investment required to maintain it. The Wall Street

Data Status✓ Full text extractedRead Original (動區 BlockTempo)

🔍Historical Similar Events· Keyword + Asset Matching6 items

2026-05-20

Wall Street Journal: OpenAI to file IPO application as early as this Friday! Targeting September listing

2026-05-19

Musk loses lawsuit against OpenAI! $134 billion damages claim falls through, biggest legal obstacle to Altman's IPO push lifted

Similarity 140%關鍵字 ipo/openai同分類 zh

2026-05-13

Anthropic valuation skyrockets to $900 billion, surpassing OpenAI! Reportedly planning to raise $30 billion, with IPO potentially launching as early as October.

Similarity 140%關鍵字 ipo/openai同分類 zh

2026-05-11

On the eve of OpenAI IPO: 600 employees cash out $6.6 billion, 75 people receive the full $30 million