The annual recurring revenue of OpenAI is approaching $70 billion, and the strong acceleration in company sales is narrowing the gap between this company and Anthropic.

According to Axios, the revenue operation rate of OpenAI has increased by over 70% since the beginning of the third quarter, and corporate sales have more than doubled since July. This growth has brought new momentum to the developers of ChatGPT; previously, Anthropic was leading in terms of disclosed revenue growth.

This change is particularly important, as corporate clients are becoming one of the core battlegrounds in the AI competition.

The business of OpenAI is accelerating.

OpenAI is increasingly shifting its business focus from consumer-oriented ChatGPT subscriptions to enterprise AI spending, covering enterprise products, API, and developer tools.

This strategy seems to be working.

Just a few weeks ago, the revenue operation rate of Anthropic reached approximately $65 billion, surpassing OpenAI, and further highlighting the advantages of Claude among corporate clients.

Today, OpenAI is rapidly narrowing this gap.

This change also occurred at a time when price competition in the AI industry has intensified. The decrease in Token costs has prompted OpenAI and Anthropic to engage in an even broader AI price war, forcing both companies to control infrastructure costs while expanding usage.

IPO Competition brings more pressure

As both companies prepare for potential listings, the revenue race becomes even more important.

OpenAI has already begun to advance the application for IPO, while Anthropic submitted its documents secretly earlier this year and is considered to have a valuation that may approach 2 trillion US dollars.

If they go public, investors will be able to gain a more in-depth understanding of the financial conditions of both companies, including a metric that is currently not very transparent: cost.

This is very important, because a significant increase in revenue does not necessarily translate automatically into profit.

Training cutting-edge models, running inference, and ensuring sufficient computing power capacity all require billions of dollars in investment each year. The latest information disclosed by Anthropic has shown the potential scale of these commitments, while OpenAI provides much less detail regarding its current expenditure base.