In a rare internal meeting, OpenAI's CFO disclosed that the incremental annualized recurring revenue (ARR) in July alone exceeded the total for the entire second quarter, driven by the explosive adoption of GPT-5.6, enterprise agents, and the coding tool Codex. Some analysts suggest this could push annual ARR to nearly $60 billion, narrowing the gap with Anthropic. At a pivotal moment with a valuation of $852 billion and a confidential IPO filing already submitted, this accelerating revenue curve could become the strongest card in its roadshow.
OpenAI is boosting its revenue curve at a staggering pace. The Chief Financial Officer of the world's most influential AI company revealed in an internal meeting that the company's incremental annualized recurring revenue (ARR) in July alone exceeded the entire second quarter, signaling an acceleration in its commercialization process.
On Thursday, according to partial transcripts of an internal meeting obtained by CNBC, OpenAI CFO Sarah Friar told employees at an all-hands meeting on Wednesday that the company's ARR in July surpassed the entire second quarter. "And the second quarter itself was quite impressive," Friar added. Board Chairman Bret Taylor also attended the meeting.
Previously, OpenAI's revenue data released in April showed an annualized recurring revenue (ARR) of $25 billion for 2026. Estimates suggest that the ARR for the second quarter (April-June) was between $25 billion and $42 billion, meaning the second quarter added a net total of approximately $17 billion in ARR.
Adding to the existing $42 billion at the end of June, OpenAI's ARR by the end of July was close to $60 billion, narrowing the gap with Anthropic, which, according to TickerTrends, had an ARR of around $74.1 billion in 2026.
This statement comes as OpenAI faces intense competition from Anthropic and a wave of low-cost open-source models. The company is actively signaling business health to employees while building valuation support for a potential massive IPO.
Growth Drivers: GPT-5.6, Enterprise Agents, and Codex
Friar and Taylor attributed the current growth momentum to three main drivers: the launch of the GPT-5.6 series, the new enterprise AI agent product ChatGPT Work, and the rapid adoption of the AI coding tool Codex.
Taylor acknowledged at the meeting that OpenAI had fallen behind Anthropic in the coding market and needed to catch up. However, he expressed encouragement about Codex's growth trajectory. "You'll see that those heavy users of Claude Code end up with hefty bills and start looking for alternatives," Taylor said.
According to a report by The Information in March, citing sources familiar with the matter, OpenAI's annualized revenue had already surpassed $25 billion at that time. Since then, with the explosive growth of Codex, the revenue curve has steepened significantly.
Competitive Pressure: Anthropic's Valuation Surpasses, But Gap Narrows
Behind OpenAI's accelerated growth is an increasingly fierce market battle. Anthropic surpassed OpenAI in valuation earlier this year and announced in May that its revenue run rate had exceeded $47 billion—while the company's full-year revenue for 2025 was about $10 billion. Anthropic's Claude Code tool has rapidly gained popularity among developers, seen as the core driver of its revenue surge.
At the same time, open-source models from China continue to apply pressure. Earlier this month, China's Moonshot AI released Kimi K3, claiming it outperforms OpenAI and Anthropic's flagship systems on some benchmarks and can be accessed at a lower cost.
Taylor acknowledged at the meeting that Anthropic had a strong start to the year but emphasized that OpenAI is regaining the initiative.
IPO Prospects: $852 Billion Valuation Awaits Market Test
OpenAI, currently valued at $852 billion, is under pressure to deliver performance matching that figure. The company filed a confidential IPO application with the U.S. Securities and Exchange Commission (SEC) in June, alongside Anthropic, but neither has disclosed a specific timeline for going public.
In terms of infrastructure investment, OpenAI disclosed to investors in February that it plans to spend about $600 billion cumulatively on computing power by 2030. According to a CNBC report earlier this week, the company is currently in talks with Nvidia for up to $250 billion in funding support to lease a large AI data center in Ohio.
To support these massive infrastructure expenditures, OpenAI needs to continuously expand its base of enterprise and developer users to generate sufficient cash flow from revenue. The July revenue data could become one of the most compelling arguments in its IPO roadshow.
