Anthropic Overtakes OpenAI in Business Customer Count: What the Numbers Show

For the first time since businesses began seriously adopting generative AI, Anthropic has pulled ahead of OpenAI in the number of verified business customers. The shift appears in fresh data from the Ramp AI Index, released as overall business AI adoption climbed to 47.6% in February 2026, up from 46.8% in January.

Anthropic captured 37% of trackable business spending on generative AI software in the first quarter, compared to 33% for OpenAI. Among the businesses tracked by Ramp, 24.4% were using Anthropic services in February, while OpenAI saw a 1.5% monthly decline — the largest single-month drop for any AI model company Ramp has tracked to date.

What the Market Shift Reveals

When enterprises evaluate both platforms side-by-side, Anthropic wins roughly 70% of the time among businesses buying AI tools for the first time. That success rate points to genuine momentum.

The picture becomes more interesting when you look at industry patterns. Anthropic currently leads in three sectors: information services, finance and insurance, and personal services. These industries tend to be more cautious about adopting new technology, and they care deeply about compliance and risk management. That suggests Anthropic's emphasis on AI safety and explainability — meaning how and why the model makes a decision — is resonating with procurement and risk teams in those fields.

Here is a wrinkle: roughly 79% of Anthropic customers also pay for OpenAI services. Meanwhile, 16% of all businesses tracked by Ramp use both providers at the same time. This dual-sourcing pattern — maintaining relationships with multiple AI vendors — appears to be how many enterprises are currently thinking about AI strategy. Rather than bet everything on a single provider, they are keeping options open.

The Broader Picture on Adoption

Overall enterprise AI adoption is accelerating, with 47.6% of businesses now using some form of AI tool. That represents nearly half of all companies tracked by Ramp, which is a striking penetration rate for technology that only entered the mainstream less than two years ago.

OpenAI's monthly decline, while notable, does not signal a wholesale exit. OpenAI still powers the largest number of business customers overall. The 1.5% drop appears concentrated among newer or smaller accounts rather than major enterprise departures. In other words, their installed base has not eroded, but their new customer pipeline has slowed.

When you look at spending patterns, the picture shifts. Anthropic's 37% share of trackable business spending, compared to their customer count, suggests they are signing larger contracts with individual businesses or gaining deeper use across multiple departments within those companies. This could mean enterprises are finding more ways to use Claude, or they are paying premium prices — or both.

Why These Shifts Matter

Different sectors favor different platforms for practical reasons. Finance and insurance companies, which move cautiously on technology, appear to prefer Anthropic's approach to safety and transparency. Information services companies may respond to Claude's reputation for factual accuracy and fewer "hallucinations" — cases where an AI invents information that sounds plausible but is wrong. Personal services companies might value how Claude handles conversation and customer interaction.

I want to flag something worth noting: we have seen this competitive pattern before, when cloud computing took off roughly a decade ago. Amazon Web Services dominated initially through first-mover advantage and technical sophistication. But Microsoft Azure and Google Cloud gained ground by solving specific enterprise problems — meeting compliance requirements, fitting into existing systems, and addressing industry-specific needs. The AI model market appears to be following a similar path, where success comes from meeting specialized enterprise needs rather than from being universally the best.

What This Means Looking Forward

The competitive landscape carries practical implications for how enterprises think about AI. If your organization is evaluating AI platforms for the first time, the data suggests you should expect stronger pitch from Anthropic in head-to-head comparisons, especially if your industry is regulated or if your use case demands that you understand how the AI reached its answer.

For enterprises already built on OpenAI infrastructure, the market dynamics point to a shift. If you have not evaluated Anthropic as a secondary option, you increasingly may be the outlier. Maintaining relationships with multiple AI providers appears to be becoming standard practice for managing risk and making sure you are using the right tool for each job.

Looking ahead, the AI model market seems to be moving toward a multi-vendor landscape, where different providers win in different domains rather than one vendor dominating everything. Enterprises should prepare for platform selection becoming a case-by-case decision rather than an organization-wide choice.