Guozhen AIGlobal AI field notes and model intelligence

Realtime AI News

OpenAI Enterprise Sales Now the Majority of Revenue as Brockman Signals 2027 IPO Readiness

OpenAI president Greg Brockman said enterprise sales now generate the majority of the company's revenue, and answered that he thinks OpenAI will be ready for an IPO in 2027. The remark, reported by Reuters, moves past OpenAI's March target of reaching parity with consumer revenue by the end of this year.

Published

OpenAI president and co-founder Greg Brockman said enterprise sales now generate the majority of the company's revenue, according to a Reuters report published on November 5. He made the remark on November 4 at a conference in Napa, California, in a session titled OpenAI's Race to Scale, moderated by Wall Street Journal reporter Keach Hagey.

Asked in the same session whether OpenAI would be ready for an initial public offering in 2027, Brockman answered that he thinks so. The reply came with no timetable, no underwriting arrangements and no valuation guidance, but against a backdrop of delayed listing plans and accelerated fundraising it stands as the clearest IPO signal the company has given.

The contrast with OpenAI's March 31 funding announcement is stark. At the time the company said enterprise accounted for more than 40 percent of revenue and was on track to reach parity with consumer revenue by the end of 2026, with the business generating about $2 billion a month. Enterprise has shifted from a business chasing the consumer side to the larger contributor.

The consumer base remains the foundation underneath that. The March announcement reported more than 900 million weekly active ChatGPT users and over 50 million subscribers, framing everyday familiarity with ChatGPT as the route into workplace adoption.

At the product level, Codex is the visible instrument of the enterprise push. OpenAI said the coding agent's weekly user base passed 2 million after growing fivefold in three months, with activity up more than 70 percent month over month. Brockman placed Codex at the centre of his account of competition for business customers.

The other half of the IPO question is money and compute. The March announcement described a round with $122 billion in committed capital at an $852 billion post-money valuation, anchored by Amazon, NVIDIA and SoftBank with continued participation from Microsoft, plus a revolving credit facility expanded to roughly $4.7 billion that remained undrawn at close.

The competitive calendar is pushing the issue too. Reuters reported that Anthropic is likely to go public before the end of 2026, which forms the reference point for Brockman's answer. For enterprise buyers, the head-to-head contest in coding and office work will not end with whichever company lists first.

One caveat matters: the majority-of-revenue claim is Brockman's verbal characterization, not a published financial disclosure, and OpenAI has not broken out enterprise versus consumer revenue. What to watch next is whether a future funding round or listing document supplies that split, and whether Codex can convert user growth into durable contracted revenue.

Why it matters

Enterprise overtaking consumer revenue changes OpenAI's valuation story from a subscription product to a contracted B2B supplier, which is the shape public-market investors tend to reward.

OpenAIBusinessIPOEnterprise
Back to realtime news

Nearby Updates

All

11/05, 10:01

基于本体+知识+Harness+Datat Agent的智能体完整落地链路!

基于本体+知识+Harness+Datat Agent的智能体完整落地链路!. Data + AI 技术实践巡展 · 上海站

11/05, 09:59

GPT-Image-2.6 Rolls Out Everywhere With Native GIF, SVG and a Separate Complexity Dial

OpenAI has fully rolled out GPT-Image-2.6 across ChatGPT, Codex and the API, cutting generation time by roughly half — six seconds at the fastest — while adding native GIF and SVG output. The API now separates a complexity parameter for scene planning from a quality parameter for rendering, paired with two variants, Flare and Sunburst.

11/05, 09:54

OpenAI's unreleased security model found 948 Linux kernel privilege-escalation flaws

OpenAI has disclosed an automated security exercise in which an unreleased cybersecurity model, wired into an agent pipeline, found 948 independent local privilege-escalation vulnerabilities in the Linux kernel for roughly $1.2 million in API spend. Discovery, validation and working exploit development were all handled by agents, at an average cost of about $1,279 per flaw.

11/05, 09:26

Apple reportedly trims internal AI budgets, cutting engineering caps from $300 to $200 a day

Apple has reportedly cut the daily AI spending cap for its engineering teams from $300 to $200, with some allowances heading to $100 a day and non-engineering groups facing limits as low as $25. The move marks a retreat from this year's tokenmaxxing push and shows that even the richest AI adopters are now budgeting model usage like any other line item.