SAN FRANCISCO — OpenAI's enterprise business now generates more revenue than its consumer subscription unit—a threshold the company's own finance chief had forecast would not arrive until the end of 2026. CFO Sarah Friar delivered that news directly to current shareholders on Aug. 14, 2026, at a meeting held the day after revenue chief Denise Dresser left the company.
Friar told attendees that OpenAI entered this year with a 60-40 split in favor of consumer revenue. That ratio has since reversed. "The majority of our revenue is now enterprise," she said, according to a person present at the meeting who asked not to be named because of confidentiality obligations.
The company's annualized revenue run rate has reached $40 billion. Total revenue grew 20 percent month-over-month in July alone. The enterprise segment outpaced that figure: business customer revenue rose 32 percent in the same period, according to presentation slides viewed at the meeting.
The $40 billion run rate marks a sharp acceleration. Earlier this year, Friar had told investors that consumer and enterprise revenue would reach parity by December 2026—a target the company hit at least two quarters early. The milestone shifts how OpenAI's revenue profile reads to any prospective investor ahead of a public offering, moving the company from a consumer subscription story toward an enterprise software one, where contract visibility and renewal rates typically command higher valuation multiples.
Friar credited a change in how corporate customers deploy AI. She described the end of what she called "tokenmaxxing"—a period when companies let employees accumulate large AI usage bills without requiring measurable output in return. Enterprise buyers have shifted to measuring cost per unit of intelligence instead. To capture that demand, OpenAI has cut prices across its model suite and improved efficiency: its newest model is 54 percent more efficient on agentic coding tasks than its predecessor, Friar said at the meeting.
The investor session was planned before this week's executive turnover but was held against a backdrop of three high-profile departures in eight days. Brad Lightcap, who spent eight years at OpenAI, announced he was leaving to start a new company two days before the meeting. Dresser stepped down the day before it, eight months after joining. She had spent more than a decade at Salesforce and most recently served as CEO of Slack before joining OpenAI as revenue chief. She described her exit in a LinkedIn post as a decision to pursue other opportunities.
OpenAI president and co-founder Greg Brockman joined Friar at the Aug. 14 session. Brockman credited Dresser for building the company's enterprise foundation, according to the person in attendance, and said he was enthusiastic about her replacement. The incoming revenue chief is Dali Rajic, who served as chief operating officer at cybersecurity firm Wiz before Google completed its acquisition of that company. Rajic was introduced to OpenAI through Josh Kushner, founder of Thrive Capital, according to a person with knowledge of the recruiting process who asked not to be named because he was not authorized to discuss it.
Executives fielded questions at the meeting about open-source Chinese AI models and whether they threaten OpenAI's pricing power. Brockman dismissed the concern. He said there is a widespread misunderstanding about open-source models being cheaper to operate in practice, though neither he nor other executives provided specific cost comparisons during the session.
Shareholders also pressed executives on the timing of an initial public offering. The executives declined to answer in detail, citing a confidential SEC filing. OpenAI has not publicly disclosed the filing or its contents.
The revenue composition shift matters structurally. Enterprise software contracts carry higher average deal sizes, longer terms and more predictable renewal cycles than month-to-month consumer subscriptions. OpenAI's consumer flagship, ChatGPT, drove the company's initial growth but generates revenue primarily through individual subscriptions at $20 per month for the Plus tier. Enterprise agreements, which include API access, custom deployments and dedicated account management, typically run into the hundreds of thousands or millions of dollars annually per customer. A company deriving the majority of its revenue from enterprise contracts presents a different risk profile to underwriters and public market investors than one dependent on consumer retention.
The 32 percent month-over-month enterprise growth rate in July, if sustained, would imply a doubling roughly every three months. Friar did not offer guidance on whether that pace continues, and OpenAI has not filed a public prospectus. The confidential SEC filing the executives referenced is consistent with a company evaluating a public offering without triggering full disclosure obligations—a common step before a formal S-1 filing.
