OpenAI grew its quarterly revenue about 50 percent to $10 billion, yet the company managed to post an adjusted operating loss of $10.9 billion over the same three months. A year earlier, the company brought in $3.9 billion, meaning top-line intake nearly tripled while operational deficits expanded in lockstep. The question facing corporate leaders tracking AI infrastructure economics is straightforward: can aggressive price cuts and enterprise repositioning actually bridge an eleven-billion-dollar quarterly chasm before an initial public offering?
The central tension lies in the disconnect between gross commercial intake and operating leverage. While OpenAI expanded its top line, its adjusted operating margin reached negative 109 percent, an improvement from negative 150 percent in Q2, though still worse than the negative 97 percent recorded a year earlier. Stock-based compensation added another $2.6 billion to expenses. Expanding sales volume has failed to resolve systemic cost inefficiencies, leaving the business model exposed to rivals running leaner operations.
The Efficiency Gap
The contrast becomes stark when measuring OpenAI against its primary competitor. Anthropic grew revenue 47 percent to $17 billion in Q3, bringing in $7 billion more than OpenAI while running an operating loss of just $567 million, a figure that already includes stock compensation.
Anthropic focuses heavily on enterprise software tools such as Claude Code and the Claude Cowork agent. OpenAI splits its balance sheet across foundational robotics research and consumer products, forcing high compute overhead across multiple fronts.
Anthropic generated $17 billion in quarterly revenue with an operating loss of just $567 million, highlighting the operational divide between targeted enterprise tooling and broad consumer deployments.
With Anthropic planning its initial public offering next week, the performance spread establishes a direct benchmark for how enterprise AI revenue converts into operating margins.
Price Cuts and Enterprise Repositioning
To close the enterprise market gap, OpenAI reshuffled its operational priorities earlier this year. The company discontinued projects like a video app, reallocated resources into its coding tool Codex, and hired Dali Rajic as its new head of sales. Simultaneously, OpenAI slashed prices on its models, capitalizing on corporate demand as third-party data showed organizations looking for cheaper alternatives to Claude.
These price concessions accelerated enterprise adoption. As Greg Brockman stated, the momentum marks an incredible revenue turning point, pointing to major corporate accounts like JPMorgan Chase. According to Brockman, enterprise contracts doubled in Q3 and now account for the bulk of OpenAI's quarterly revenue.
One counter-argument holds that OpenAI's massive consumer ecosystem will eventually absorb infrastructure costs as monetization matures. ChatGPT now counts 1.2 billion weekly active users and 100 million paying subscribers after picking up 22 million new sign-ups. Furthermore, the new ChatGPT advertising product reached an annualized revenue run rate of $1 billion in August.
Commercial subscriptions and a billion-dollar ad run rate demonstrate immense scale, yet these gains have not stopped adjusted quarterly operating losses from widening past $10 billion.
Consumer scale provides massive user engagement, but the compute expenditure required to serve 1.2 billion weekly users continues to compound operating shortfalls. This fundamental compute overhead explains why revenue growth to $10 billion runs parallel to widening losses: every new active consumer query scales infrastructure expenses faster than subscription yields.
Both providers are confronting a tech sell-off across public financial markets. Anthropic had to cut its IPO valuation target significantly ahead of its public debut next week. OpenAI is targeting an IPO next year and is attempting to raise fresh capital at a $1.25 trillion valuation, though it has so far secured less than half of its planned $30 billion target. OpenAI is asking public and private markets to price it at $1.25 trillion for generating $10 billion in quarterly sales, while its closest competitor generates $17 billion at a fraction of the operating burn.
How the picture shifted
Shutting down the video application and slashing model prices secured enterprise deals, but servicing over a billion weekly users burns infrastructure faster than subscriptions can replenish it. The diverging priorities show up against Anthropic, which targeted software tools and kept its operating deficit to $567 million while bringing in more revenue. Every unpaid prompt pushes the business closer to an IPO while methodically burning billions of dollars on server racks.