A New Leader in Corporate AI Spending
Corporate AI spending patterns just delivered a symbolic turning point in enterprise AI adoption. According to Ramp’s latest AI Index, Anthropic captured 34.4% of corporate AI spending tracked on its expense platform, nudging ahead of OpenAI’s 32.3% for the first time. This data is grounded in actual credit card and invoice transactions from more than 50,000 businesses, not surveys or self-reported intent, making it a concrete measure of paid adoption. While OpenAI still commands massive consumer mindshare through ChatGPT and remains a revenue heavyweight, the Ramp numbers highlight a different race: which vendor is winning when finance teams authorize real subscriptions for business AI tools. Anthropic’s gain of 3.8 percentage points, alongside OpenAI’s 2.9-point decline over the same period, signals that enterprises are reevaluating which AI partner best supports day-to-day work, not just experimentation.
From Model Showcases to Workflow Integration
The Anthropic vs OpenAI contest in corporate AI spending increasingly reflects a shift in what buyers value. Early enthusiasm for frontier models was driven by eye-catching demos and general-purpose chatbots. Now, procurement teams are prioritizing AI that integrates smoothly into existing workflows, supports repeatable tasks, and feels reliable enough to anchor long-term processes. Businesses are not simply choosing the most famous brand; they are choosing tools that reduce friction and fit how teams already operate. Anthropic’s positioning leans heavily into this workflow orientation. Claude is praised less for personality and more for its ability to maintain context, reason through complex, messy material, and plug into recurring internal processes. As half of the companies in Ramp’s dataset now pay for AI tools, AI is moving from experimental side project to embedded infrastructure, and the vendors winning these contracts are those that treat enterprise AI adoption as a workflow challenge rather than a capabilities beauty contest.
Winning Over Technical Teams First
Anthropic’s rise has been powered by a deliberate focus on demanding, technical customers. Ramp’s data shows Anthropic leading in information, finance, and professional services—sectors where enterprise AI adoption is already high and quality expectations are unforgiving. These teams rely on AI for code generation, research, analysis, drafting, and review, where small errors can cascade into major time costs. For them, a model’s ability to remain consistent, handle long contexts, and support complex reasoning is more important than conversational flair. Claude Code has emerged as a serious tool for developers, reinforcing Anthropic’s reputation among power users. This aligns with Ramp’s observation that Anthropic started with very technical buyers, executed well, and then broadened outward with workflow-oriented products like Cowork. In practice, that means the first AI stack a startup or finance team adopts often becomes their default—shaping documentation, approvals, and automation, and making later switching both technically and culturally expensive.
What Ramp’s Data Reveals—and What It Misses
Ramp’s AI Index offers a rare, transaction-level view into business AI tools, but it comes with important caveats. The dataset primarily captures discrete AI purchases made via corporate cards and standard SaaS-style subscriptions. Massive enterprise contracts, bespoke deals, and AI services bundled into broader cloud agreements often sit outside Ramp’s visibility. This helps explain why some major providers appear smaller than expected in the index, and why the Anthropic vs OpenAI numbers should be read as a lens on unbundled AI buying decisions rather than the entire market. Still, the signal is meaningful: when companies consciously pick and pay for specific AI vendors instead of defaulting to existing bundles, Anthropic now leads. It also shows how volatile this market remains. Ramp notes that OpenAI experienced its largest single-month share drop in February, demonstrating how shifts in pricing, packaging, or developer tools can rapidly move spend as businesses refine their enterprise AI adoption strategies.
The Next Phase of Enterprise AI Competition
Anthropic’s lead in corporate AI spending is not an insurmountable moat, but it is a critical milestone as the market consolidates around business AI tools that deliver dependable work. OpenAI still holds advantages in consumer recognition, developer ecosystems, and major enterprise relationships, and has already responded by courting business users with more aggressive coding offerings and bundles. This underscores a broader reality: model quality and brand alone are no longer sufficient differentiators. As companies experiment with agentic AI—tools that can autonomously run tests, search files, and chain actions—predictable pricing, rate limits, and governance will matter as much as raw performance. The durable advantage may accrue to vendors that align with how organizations actually operate: integrating into workflows, offering transparent economics, and sustaining trust over time. For buyers, the lesson is clear: treat AI procurement like any critical system decision, testing vendors against real work rather than hype.