Business AI race tightens as OpenAI gains on Anthropic

Ramp data indicates OpenAI is growing faster than Anthropic among its U.S. business customers so far in Q3, though Anthropic still holds the lead in share. The numbers suggest enterprise AI spending is expanding, but provider loyalty may remain fragile.

Business AI race tightens as OpenAI gains on Anthropic

OpenAI is showing fresh momentum with U.S. business users, according to new data from Ramp, the corporate credit card and expense management company. Anthropic still leads among Ramp’s paying business users, but the latest trend points to a more unsettled enterprise AI market than a simple winner-takes-all story.

OpenAI Is Gaining, But Anthropic Still Leads

Ramp’s data shows that OpenAI has begun gaining on Anthropic among U.S. businesses using Ramp’s bill pay and corporate card products. The shift matters because OpenAI had once been the clear leader with both business and consumer users, before losing the lead among Ramp’s paying business users in May.

In May, Anthropic reached 41% market share among these Ramp customers, while OpenAI stood at 39%. OpenAI has not taken the lead back since then. As of July, Anthropic had nearly 44%, while OpenAI had nearly 40%.

That means the current picture is mixed. Anthropic remains ahead in this slice of the market, but OpenAI’s recent growth suggests the gap is not fixed. Ramp economist Ara Kharazian said the most recent data shows OpenAI growing faster than Anthropic in Q3 to date among this customer segment.

The quarter is not over, and the source notes that there is still a month left. In a market where each new model release can quickly shift customer behavior, the trend could change again before Q3 closes.

What Ramp’s Data Can And Cannot Show

The data covers more than 70,000 American businesses that spend billions through Ramp’s bill pay and corporate card products. That is a large pool of business spending behavior, and it offers a useful signal about how companies are paying for AI tools.

At the same time, it is not a complete picture of the entire business AI market. Ramp’s customers operate across industries, but they skew toward the tech industry because Ramp is a popular Silicon Valley corporate credit card. The data also excludes large enterprises that use spend-management products from providers like American Express instead of Ramp.

Ramp also declined to provide actual dollars spent. The company shared percentages, which can show direction and relative movement, but not the full scale of revenue flowing to each AI company.

Those limits are important. The numbers should not be read as a final ranking of OpenAI and Anthropic across all enterprise customers. They are better understood as market indications from a meaningful but specific business-spending dataset.

Model Releases May Be Moving Business Buyers

The Ramp data suggests business AI customers may be willing to move between providers as new models arrive. That is a key implication for both OpenAI and Anthropic, because it raises questions about how durable enterprise AI spending really is.

Kharazian connected OpenAI’s recent growth to its newer model performance. He posted on X:

GPT-5.6 Sol is really good, increasingly the choice for developers

He also said:

Fable 5, meanwhile, disappointed both in adoption and real-world application given price + data retention requirements imposed by regulators

The source cautions that this may be an oversimplification. Fable, Anthropic’s higher-end model tier, is expensive, but it is also designed for a more targeted group of use cases than a general chatbot.

Still, one customer concern around Anthropic did stand out. Anthropic caused some outrage when it warned Fable users that it must retain their data for 30 days. For business users, requirements around data handling can shape buying decisions, especially when companies are weighing cost, model performance, and internal risk.

The AI Spending Pool Is Still Expanding

The competitive story is not only about OpenAI taking share from Anthropic or Anthropic defending its lead. Ramp’s data also suggests that both companies should be growing business revenue because the overall market among these customers is expanding.

The share of companies paying for AI among Ramp customers has been climbing steadily. It topped 50% in March and reached nearly 56% by July.

That growth changes how the competition should be read. A company can lose some relative share while still benefiting from a larger pool of paying business customers. Likewise, a company gaining share may be growing inside a market that is already moving upward.

For investors and business software buyers, the bigger lesson is volatility. Anthropic has not won permanently, and OpenAI has not disappeared from the business market. The data points instead to a market where customers keep reassessing tools as pricing, model quality, application fit, and data requirements change.

Why This Matters For Enterprise AI

Until OpenAI and Anthropic are close enough to their planned IPOs to release financials, outside data sources will remain important for understanding business momentum. Ramp’s dataset does not answer every question, but it offers a concrete view into how many businesses are paying for AI and which providers are gaining share inside that group.

The clearest takeaway is that enterprise AI spending is not yet settled. Anthropic leads OpenAI among Ramp’s paying business users as of July, while OpenAI is currently growing faster in Q3 to date. The market itself is also expanding, which means the competition is happening on a larger base of paying companies.

For now, the business AI race looks less like a stable hierarchy and more like an active contest. OpenAI is gaining ground, Anthropic remains ahead in Ramp’s data, and customers appear ready to reconsider their choices as the products change.