OpenAI Enterprise Revenue Surpasses Consumer as Run Rate Hits $40 Billion

The Numbers Behind the Crossover

OpenAI’s annualized revenue run rate has reached $40 billion. More notably, enterprise revenue has overtaken consumer revenue — a reversal that Friar said arrived ahead of schedule.

Earlier this year, Friar had projected the two segments would reach parity by the end of 2026. Instead, the lines crossed sooner. “We entered the year at 60-40, but enterprise has accelerated much faster than expected and those lines have now crossed,” she told investors, according to a person in attendance.

The underlying growth figures reinforce the point. OpenAI’s run rate grew 20% month over month in July. Business customers grew even faster — up 32% in the same period.

What Drove Enterprise Acceleration

Friar pointed to a behavioral shift among enterprise customers as a key factor. The era of what she called “tokenmaxxing” — where companies allowed employees to accumulate large AI bills without demonstrating proportional output — appears to be over.

Customers have moved toward measuring cost per unit of intelligence rather than raw usage volume. That shift favors providers who can demonstrate efficiency gains alongside capability improvements.

Friar highlighted that OpenAI’s newest model is 54% more efficient on agentic coding tasks, and noted recent price reductions across its model suite. For enterprise buyers managing AI spend at scale, those numbers matter more than benchmark rankings.

A Week of C-Suite Turbulence

The investor meeting was planned in advance but landed at an awkward moment. It followed the departure of revenue chief Denise Dresser, who stepped down after eight months in the role. Two days before that, longtime executive Brad Lightcap announced he was leaving after eight years at the company.

OpenAI president and co-founder Greg Brockman joined the meeting and acknowledged Dresser’s contributions, specifically crediting her with building the enterprise foundation. Her replacement, Dali Rajic — previously chief operating officer at cybersecurity firm Wiz — was introduced to OpenAI through Thrive Capital founder Josh Kushner.

The executive turnover adds some uncertainty to an otherwise strong revenue story. Whether Rajic can sustain the enterprise momentum Dresser helped build is a reasonable question for anyone evaluating OpenAI’s trajectory.

Open Source, IPO, and Advertising

Investors raised questions about competitive pressure from open-source Chinese models. Brockman pushed back, suggesting there is a widespread misunderstanding about open-source models being inherently cheaper to operate at scale.

On the IPO question, executives declined to comment, citing a confidential SEC filing.

The advertising angle is worth watching separately. Friar said OpenAI has made meaningful progress in advertising since beginning tests in ChatGPT in February, with ad revenue now approaching a $1 billion annualized run rate. That is a new revenue stream that did not exist at the start of the year.

What This Means for AI Tool Buyers

For teams evaluating AI tools, the enterprise-first revenue shift at OpenAI has practical implications:

• Pricing will increasingly reflect enterprise value, not consumer accessibility. Expect continued model price reductions to compete for volume contracts, not necessarily to benefit individual users.
• Efficiency metrics are becoming the sales argument. If OpenAI is leading with cost-per-unit-of-intelligence framing, expect competitors to follow. That gives buyers a more useful lens for comparison than raw capability claims.
• Agentic coding is a priority investment area. The specific callout of 54% efficiency gains on agentic coding tasks suggests where OpenAI is concentrating development resources — relevant for any team building or evaluating AI-assisted development workflows.

The $40 billion run rate is a headline number. The more useful signal is that enterprise customers are now the primary audience OpenAI is building for.