Anthropic’s Business Revenue Has Gone From Being 10% Of OpenAI’s To 200% Of OpenAI’s In The Past Year

Anthropic has recently surged in popularity across consumers with its app hitting the top of the App Store, but its surge among business users has been even more remarkable.

New data from Ramp Economics Lab, which tracks corporate card and bill pay spending from more than 50,000 U.S. businesses, shows that Anthropic’s share of business spending on AI chat subscriptions — measured against OpenAI alone — has undergone a transformation that would have seemed implausible just twelve months ago. The Ramp data excludes spend on AI models from companies other than OpenAI and Anthropic, so what it captures is the direct head-to-head battle between the two most prominent AI labs for corporate wallet share. In early 2025, Anthropic’s combined product lines — Claude Enterprise, Claude Team, and other subscriptions — accounted for roughly 10% of that combined OpenAI-plus-Anthropic spend. As of February 2026, that figure has climbed past 65%, meaning Anthropic now commands around twice the business subscription spend that OpenAI’s suite of ChatGPT products does within this two-company comparison.

anthropic vs openai business spend

The Numbers Tell A Striking Story

At the start of 2025, the Ramp data shows a two-company race that was overwhelmingly one-sided. Within the OpenAI-plus-Anthropic universe, ChatGPT Business, ChatGPT Plus, ChatGPT Pro, and other OpenAI products collectively occupied roughly 90% of combined spend. Anthropic’s presence, while not negligible, was confined to a narrow band at the bottom of the chart — Claude Enterprise carrying the largest slice, with Claude Team only beginning to register.

What followed over the next twelve months is a sustained, structural shift rather than a one-time bump. Month by month, Anthropic’s aggregate share of this two-company pie expanded — not because OpenAI’s spending fell in absolute terms, but because Anthropic’s grew dramatically faster. Claude Team, in particular, emerged as the engine of growth, its orange bar on the Ramp chart widening through the second half of 2025 and into early 2026. By January and February 2026, Anthropic’s combined products had crossed the 65% threshold within the OpenAI-Anthropic comparison, a swing of roughly 35 percentage points in twelve months.

Claude Team Is The Engine

A closer look at Anthropic’s product mix reveals that Claude Team has driven the bulk of the gain. The product — designed for small to mid-sized businesses that want shared workspaces, administrative controls, and higher usage limits — appears to have found strong product-market fit in exactly the segment of U.S. businesses that Ramp serves. Claude Max, a higher-tier offering, has also grown noticeably in the most recent months, suggesting Anthropic is beginning to capture more sophisticated enterprise users alongside the broader SMB market.

Claude Enterprise, while the longest-standing Anthropic business product, represents a smaller slice — likely reflecting the longer sales cycles and concentrated deal sizes typical of large enterprise software. The real story in the Ramp data is the democratisation of Claude adoption: businesses of all sizes appear to be choosing it, not just the Fortune 500.

Why Businesses Are Switching

Several factors appear to be driving the shift. Claude’s reputation for producing longer, more nuanced written outputs has made it a preferred tool for knowledge workers in legal, finance, consulting, and communications. Anthropic’s consistent model quality across successive releases — Claude 3.5 Sonnet in mid-2025 was widely regarded as setting a new standard for practical business tasks — built genuine loyalty among power users who then advocated for it within their organisations.

The timing of the most recent acceleration is also notable. Anthropic’s public confrontation with the U.S. government in late February 2026, which ended with the company topping the App Store for the first time and surpassing Grok and DeepSeek in web traffic, generated a wave of consumer attention. While the Ramp data tracks business invoices rather than consumer downloads, the spillover effect is real: enterprise decision-makers are also consumers, and a week of front-page coverage of Anthropic’s values and product quality almost certainly accelerated conversations that were already underway in corporate procurement meetings.

Model Performance Is Backing The Business Case

The business momentum is increasingly matched by independent validation of Claude’s technical capabilities. Programming legend Donald Knuth — Stanford computer scientist and author of The Art of Computer Programming — recently documented his reaction to Claude Opus 4.6 solving a combinatorics problem he had been working on for several weeks, calling the development a “dramatic advance in automatic deduction and creative problem solving.” Knuth is not a tech booster — he is an 87-year-old foundational figure in computer science who had previously been skeptical of AI’s depth. His endorsement signals that Claude’s capabilities are penetrating even the most demanding and skeptical corners of the expert community.

What This Means For OpenAI

OpenAI retains enormous advantages: brand recognition, a more mature enterprise sales organisation, deep integrations across the Microsoft ecosystem, and a consumer footprint built over several years. It is also worth keeping the Ramp data in its proper context: it measures AI chat subscription spend among Ramp’s U.S. business customers, head-to-head between OpenAI and Anthropic only. It excludes API spend — where OpenAI likely holds significant lead — as well as spend on Google, Microsoft Copilot, and other competitors. The total AI spend market is much larger than this two-company slice.

That said, the direction of travel within this comparison is clear, and the speed of change is striking. Twelve months ago, Anthropic was a distant second to OpenAI in the category these numbers measure. It has now pulled decisively ahead. The competitive dynamics of enterprise AI are more fluid than any single snapshot might suggest — and Anthropic, for now, has the momentum.

The Road Ahead

The more important question is whether Anthropic can hold what it has won. Business software adoption follows a stickiness curve: once teams build workflows around a tool, switching costs rise sharply. If the companies now appearing in Ramp’s data embed Claude into their operations over the next year — through API integrations, customised deployments, and organisational training — that spend becomes structurally durable.

Anthropic is entering 2026 with a product line that is resonating with business buyers, a consumer brand that has just received an unexpected boost, and a growing body of expert validation. While it’s still early days, it appears that Anthropic now has both its consumer and enterprise bases covered as it looks to take on the bigger players in the space.

Posted in AI