OpenAI is rapidly closing the gap with Anthropic in the competition for U.S. business users, according to new spending data. In the third quarter to date, OpenAI's growth rate among businesses has surpassed Anthropic's, narrowing a lead that Anthropic had established in May.
The data, compiled by financial technology firm Ramp, tracks spending across more than 70,000 U.S. businesses. As of July, Anthropic held a 43.5% share of paying business customers on Ramp's platform, a slight increase from the previous month. OpenAI's share reached 39.7% in July, showing a smaller gain over the same period. However, Ramp economist Ara Kharazian noted that OpenAI's expansion has been at a faster rate in the current quarter.
This dynamic suggests that enterprise AI spending remains fluid, with businesses not exhibiting strong loyalty to a single provider. Companies are actively experimenting with various AI models, leading to shifts in market share. The overall share of businesses paying for AI services on Ramp's platform has expanded, rising from just over 50% in March to nearly 56% in July.
OpenAI's recent momentum is attributed to the strong adoption of its GPT-5.6 Sol model by developers. In contrast, Anthropic's newer Fable 5 model has seen weaker uptake. Ramp data indicates that Fable 5 accounted for only 6% of tokens and 11.4% of dollars spent on Anthropic models in July. This underperformance is partly due to Fable 5's higher cost, approximately double that of GPT-5.6 Sol, which has limited business appetite. Additionally, Anthropic faced backlash after informing Fable users of a mandatory one-month data retention policy, a requirement that may deter some businesses.
The competitive landscape has seen shifts in market leadership previously. In May, Anthropic surpassed OpenAI in business adoption for the first time, capturing 41% of Ramp's paying business customers compared to OpenAI's 39%. Prior to that, OpenAI had maintained a lead since the launch of ChatGPT. Anthropic's growth over the past year was significant, quadrupling its business adoption, while OpenAI saw a more modest increase.
The broader trend in enterprise AI spending highlights a search for cost-effectiveness and performance. Businesses are increasingly exploring open-source and more affordable model options, contributing to the volatility observed between major providers. This experimentation indicates that while AI adoption is growing, the long-term commitment to specific vendors is still being determined.