OpenAI Falls Short of Weekly Activity and Revenue Targets, IPO Faces Pressure

According to a Wall Street Journal report, OpenAI has recently failed to meet internally set targets for new users and revenue, raising concerns among some of the company's leadership about data center spending capabilities. CFO Sarah Friar has indicated to other executives that if revenue growth is not fast enough, the company may struggle to pay for the computing resources contracts in the future.

This pressure stems from OpenAI's highly aggressive strategy of computational power expansion in the past. The report states that Sam Altman drove the company to lock in a significant amount of data center capacity last year, incurring about 60 billion US dollars in future expenditure commitments for OpenAI, with the core logic being to use more computing power to support model iteration and product growth.

The slowing growth is changing the board's tolerance for this strategy. OpenAI originally planned to reach 1 billion weekly active users for ChatGPT by the end of last year, but the company has not yet announced this milestone, and the annual revenue target for ChatGPT has also not been achieved.

Competitive pressures are further compressing the margin of error for this model. The report mentions that Google Gemini grew rapidly at the end of last year and eroded OpenAI's market share; earlier this year, OpenAI also missed several monthly revenue targets due to lagging behind Anthropic in coding and the enterprise market.

To cushion the pressure from slowing growth, OpenAI is relying on financing, product growth, and cost control simultaneously. The company recently completed financing of 12.2 billion US dollars and stated that Codex is growing rapidly, also reducing costs by cutting projects such as Sora; on the other hand, the company stated in investor memos that compared to Anthropic, OpenAI has gained more computing capacity, which helps it reach more users.

However, this does not eliminate the constraints of cash consumption. The report states that some financing comes with conditions, and even if OpenAI achieves aggressive revenue targets, the company still expects to consume this financing over the next three years.

Therefore, the internal focus is shifting from "whether we can buy more computing power" to "whether revenue can cover the expansion of computing power" and "whether there is a governance foundation for going public on schedule". Friar has recently expressed reservations about the plan to go public within the year, citing the need for OpenAI to improve internal controls to meet the stricter disclosure requirements of publicly traded companies.