OpenAI Gaining on Anthropic in the Enterprise AI Race
Recent data from corporate spend management platform Ramp indicates that OpenAI gaining on Anthropic among U.S. business users may be shifting the dynamics of the enterprise AI market. While Anthropic currently holds the overall lead, the trend suggests that OpenAI gaining on Anthropic reflects a potentially significant momentum shift in the second half of 2026. This development carries important implications for investors, developers, and businesses evaluating their AI vendor strategies.
The Data Behind the Trend
Ramp, a corporate credit card and expense management company, released new data showing a notable shift in market share among its paying business users. The data covers more than 70,000 American businesses that spend billions through Ramp’s bill pay and corporate card products. While Ramp’s customer base spans multiple industries, it does skew toward the technology sector, given its popularity as a Silicon Valley corporate card.
According to Ramp economist Ara Kharazian, Anthropic captured a 41% market share among these paying business users in May, surpassing OpenAI’s 39% share. By July, Anthropic had grown to nearly 44%, with OpenAI holding approximately 40%. On the surface, this suggests Anthropic has maintained a comfortable lead. However, a closer examination of the most recent data reveals that OpenAI gaining on Anthropic is currently happening at a faster rate in Q3 to date.
Kharazian shared his insights on X, noting that GPT-5.6 Sol is increasingly becoming the choice for developers. He contrasted this with Fable 5, which he said disappointed in both adoption and real-world application given its pricing structure and data retention requirements imposed by regulators. While this may be an oversimplification—Fable being Anthropic’s higher-end tier built for targeted use cases rather than general chatbot functionality—it highlights a perception shift among business users.
Why OpenAI Is Gaining Momentum
Several factors appear to be driving OpenAI gaining on Anthropic in recent months. First, the release of GPT-5.6 Sol has generated positive reception among developers, who are reportedly choosing it for their projects with increasing frequency. Second, Anthropic’s Fable 5 has faced headwinds, including customer outrage when the company warned users about mandatory 30-day data retention requirements. These regulatory compliance measures, while necessary for some use cases, have created friction for businesses concerned about data privacy and operational flexibility.
It is also worth noting that Fable’s higher price point positions it for a more targeted set of enterprise use cases, making it less broadly appealing than a general-purpose chatbot. This natural segmentation may limit its adoption compared to OpenAI’s more accessible offerings. As a result, the trend of OpenAI gaining on Anthropic may reflect businesses opting for more flexible, cost-effective solutions for their everyday AI needs.
Enterprise AI Spending Volatility
One of the most significant takeaways from the Ramp data is the volatility of enterprise AI spending. Businesses appear willing to switch between AI providers as each lab releases new models, suggesting that customer loyalty in this space is far from guaranteed. This behavior should give both companies’ investors pause about how “sticky” enterprise AI spending really is.
The data indicates that Anthropic has not won permanently. Companies are open to shifting their spending based on model performance, pricing, compliance requirements, and other factors. This means that OpenAI gaining on Anthropic today could reverse with the next major model release. The AI industry moves at a rapid pace—with a quarter representing roughly 30 “AI years”—so the competitive landscape remains highly fluid.
Despite the fierce competition for market share, both companies appear to be growing their business revenue because the overall market is expanding. The percentage of companies that pay for AI among Ramp customers has steadily climbed, topping 50% in March and reaching nearly 56% by July. This suggests that the pie is growing even as OpenAI and Anthropic compete for larger slices.
Limitations of the Data
It is important to acknowledge the limitations of this data. Ramp’s figures do not represent the total enterprise AI market. The data excludes large enterprises that use spend-management tools from providers like American Express rather than Ramp. Additionally, Ramp declined to provide actual dollars spent, sharing only percentage-based market share figures. This means we lack a complete picture of revenue distribution between the two companies.
However, the data still offers a valuable window into mid-market and tech-sector AI adoption. For a segment that often serves as an early indicator of broader trends, the pattern of OpenAI gaining on Anthropic is noteworthy. It suggests that even in a market where Anthropic has established a lead, OpenAI retains significant competitive strength.
What This Means for Businesses and Investors
For businesses evaluating AI vendors, the current trend underscores the importance of flexibility and vendor diversification. Relying too heavily on a single AI provider may expose organizations to the risk of model degradation, pricing changes, or compliance issues that could disrupt operations. The volatility reflected in the Ramp data suggests that maintaining the ability to switch between providers—or use multiple providers simultaneously—may be a prudent strategy.
For investors, the data highlights both opportunity and risk. The enterprise AI market is expanding rapidly, creating significant revenue potential for both OpenAI and Anthropic. However, the lack of customer stickiness means that market leadership can shift quickly. Companies that fail to innovate or that introduce friction for their users may see their market share erode, while those that deliver compelling new capabilities can capture it.
The trend of OpenAI gaining on Anthropic also suggests that pricing and regulatory considerations are becoming increasingly important factors in enterprise AI purchasing decisions. As businesses mature in their AI adoption, they are likely to become more discerning about total cost of ownership, compliance burdens, and integration complexity.
As both OpenAI and Anthropic continue working toward their planned IPOs, their financial performance will eventually become public. Until then, third-party data like Ramp’s offering valuable glimpses into competitive dynamics. The current data suggests that while Anthropic holds a lead, OpenAI gaining on Anthropic indicates that the race remains very much alive.
The key takeaway is that enterprise AI is still in its early stages. Adoption rates among Ramp customers have climbed steadily, but even at 56%, there is substantial room for growth. As new models emerge and use cases expand, both companies have opportunities to capture additional market share. Whether OpenAI can sustain its current momentum or Anthropic will regain its footing remains to be seen, but for now, the trend of OpenAI gaining on Anthropic is one that industry observers will be watching closely.
