The enterprise AI market is flipping. New data from Ramp’s March 2026 AI Index shows Anthropic now wins roughly 70% of head-to-head matchups against OpenAI among businesses purchasing AI services for the first time.
A year ago, one in 25 businesses on Ramp paid for Anthropic. Now it’s nearly one in four.
The Numbers Tell the Story
Anthropic’s adoption hit 24.4% of businesses, with 4.9% month-over-month growth. OpenAI, meanwhile, posted a 1.5% decline—the largest single-month drop for any AI model company Ramp has tracked.
OpenAI still leads in total business adoption. But the trend line matters more than the snapshot.
Revenue tells a similar story. Anthropic hit $19 billion in annualized revenue in March, up from $9 billion at the end of 2025. That’s a doubling in roughly three months. According to Anthropic CFO Krishna Rao, “Claude is increasingly becoming critical to how businesses work.”
Claude Code Is the Engine
The product driving this growth is Claude Code, Anthropic’s AI coding assistant. It went from zero to $2.5 billion in annualized revenue in ten months and now accounts for more than half of all enterprise spending on Anthropic products.
Claude holds 42-54% of the code generation market while OpenAI sits at 21%. For developers choosing AI coding tools, Claude has become the default.
Enterprises represent roughly 80% of Anthropic’s business, per CEO Dario Amodei. Consumer users matter less to the bottom line than businesses willing to pay enterprise pricing for reliable outputs.
Why Businesses Are Switching
Enterprises aren’t just chasing benchmarks. The shift reflects different priorities:
Predictability: Business environments need repeatable, stable outputs. Models that produce wildly different results for the same prompt create operational headaches.
Data Security: Enterprises need assurance that confidential information stays protected. Anthropic’s positioning on data handling has resonated with security-conscious buyers.
Governance: Companies want guardrails they can actually enforce. The ability to set output restrictions aligned with company policies matters when deploying AI at scale.
Multi-Model Flexibility: Organizations are adopting multiple AI providers rather than betting on a single vendor. Anthropic benefits from companies hedging their OpenAI dependency.
The Ramp analysis attributes Anthropic’s success partly to “cultural positioning rather than pure performance metrics.” In practical terms: comparable products, different vibes.
OpenAI’s Challenges
OpenAI remains the largest AI provider by total adoption. But several factors are working against it:
Consumer-heavy user base means lower revenue per user. Most OpenAI users pay nothing, while Anthropic’s enterprise focus generates higher margins.
The Pentagon controversy helped Anthropic. After OpenAI’s stumbles in government contracting, some enterprises viewed Anthropic as the more stable long-term partner.
Brand perception shifted. OpenAI’s various public disputes—from the Sam Altman boardroom drama to copyright lawsuits—created uncertainty for risk-averse enterprise buyers.
What This Means
The AI market is maturing past “who has the most impressive demo.” Enterprise buyers care about reliability, security, and vendor stability. Those are Anthropic’s selling points.
None of this means OpenAI is failing. It’s still enormous and still growing in absolute terms. But the growth rate differential matters for where the market heads next.
For businesses evaluating AI providers: the market is now genuinely competitive. That’s good for pricing, good for features, and good for leverage in vendor negotiations.
For developers: Claude Code’s dominance in enterprise coding suggests it’s worth evaluating seriously if you haven’t already. The network effects of tool adoption can lock in ecosystems.
The enterprise AI race isn’t over. But Anthropic is no longer chasing.