OpenAI Prints $4.3B in Revenue—But Torches Billions in Cash in an ‘Era-Defining Money Furnace'

OpenAI’s latest financial disclosures underscore both the scale of its business and the extraordinary cost of running it. According to figures reported by The Information and highlighted by Reuters and the Financial Times , the maker of ChatGPT generated $4.3 billion in revenue during the first half of 2025, a 16% increase compared to its entire revenue haul in 2024. On a run-rate basis, the company is on track to hit its full-year target of $13 billion in sales, cementing its position as the leading commercial player in generative AI. Yet the numbers also reveal the sheer financial pressure behind its operations: cash burn in the first six months reached $2.5 billion, with a full-year target of $8.5 billion.

The primary driver of this burn is research and development. OpenAI poured $6.7 billion into R&D during the first half, largely tied to training its frontier models and maintaining the heavy cloud infrastructure required to serve ChatGPT’s global user base. These are not discretionary costs; sustaining growth in generative AI requires immense computing power, and the company continues to prioritize scale and innovation. At the same time, OpenAI’s sales and marketing spend reached $2 billion in just six months, nearly doubling all of 2024’s total and underscoring the competitive fight to win customers and talent. Stock-based compensation added another layer, reaching $2.5 billion in the half, nearly double last year’s pace.

Despite the red ink, OpenAI’s balance sheet remains fortified. The company closed June with roughly $17.5 billion in cash and securities, providing a large enough cushion to continue executing on its ambitious roadmap. That liquidity also gives OpenAI the flexibility to expand infrastructure and fend off rivals such as Anthropic and Google DeepMind, who are also scaling aggressively. The burn rate, however, is drawing attention to whether the company can manage a sustainable balance between innovation and financial discipline.

On the revenue side, the growth story remains compelling. ChatGPT subscriptions, API licensing, and enterprise partnerships with Microsoft’s Copilot suite remain the backbone of income. ChatGPT alone delivered $2.7 billion in 2024, and new features such as embedded commerce tools are helping diversify streams. Partnerships with enterprise customers are broadening adoption across industries from healthcare to finance, while OpenAI’s push into international markets—such as its planned expansion into India by year-end—suggests further upside.

A critical structural element in OpenAI’s financial model is its revenue-sharing agreement with Microsoft. The company is required to remit 20% of revenue to its strategic partner, a figure that is projected to decline over time, saving OpenAI as much as $50 billion through 2030. This dynamic highlights both the strength and complexity of the relationship: Microsoft is OpenAI’s largest distribution partner and investor, but also a shareholder that captures a significant portion of its economics.

The valuation story continues to attract attention. OpenAI is reportedly preparing a tender offer that would allow employees to sell stock at a valuation near $500 billion, up sharply from $260 billion earlier this year. Such a figure would put OpenAI among the most highly valued private tech companies in history and highlight the degree of investor conviction in the company’s long-term role in the AI ecosystem. However, critics note that valuation momentum is coming alongside an “era-defining money furnace,” as the Financial Times described it, pointing to the tension between revenue growth and mounting losses.

Support from hardware partners underscores the scale of OpenAI’s ambition. Nvidia has pledged to invest as much as $100 billion into the company, providing high-performance data center chips to underpin training and inference at scale. This aligns OpenAI’s growth trajectory with Nvidia’s own hardware roadmap, emphasizing the symbiotic relationship between model developers and chipmakers. SoftBank has also expressed willingness to commit tens of billions toward OpenAI’s data center buildout. These moves highlight how capital-intensive generative AI has become, as well as the network of stakeholders betting on OpenAI’s leadership.

The results matter well beyond the company itself. OpenAI is not publicly traded, but its financial disclosures have broad market implications. The firm’s revenue growth validates the demand for generative AI services, offering support for the valuations of peers such as Anthropic, Cohere, and even Google’s DeepMind unit. At the same time, its burn rate highlights the risks, reminding investors that scaling these technologies requires unprecedented capital intensity. For semiconductor firms like Nvidia, AMD, and Broadcom, OpenAI’s spending spree translates into massive demand for chips and infrastructure. For cloud providers such as Microsoft and Amazon, it reinforces the opportunity to monetize AI workloads.

Still, OpenAI’s trajectory raises questions for the broader AI sector. Will customers sustain willingness to pay for subscriptions and enterprise integration at the pace needed to offset costs? Can the company manage talent and infrastructure spending without outstripping even its immense funding base? And how much investor patience will there be if the company’s profitability timeline stretches further into the future?

For now, the results present a dual narrative: OpenAI is proving its dominance in AI commercialization while burning through capital at a historic rate. The balance between those two forces will shape not just the company’s future, but also the fortunes of the AI industry at large.