Claude Tops OpenAI in Enterprise AI for the First Time: What the Ramp Billing Data Shows
TL;DR
For the first time, Anthropic's Claude holds more US enterprise AI spend than OpenAI — 34.4% vs 32.3%, per the May 2026 Ramp AI Index tracking 50,000+ companies. Claude Code is the main driver behind Anthropic's 4x year-over-year enterprise growth.
A billing dataset tells you something a product launch can’t fake.
The May 2026 Ramp AI Index, drawn from spending and invoice records at more than 50,000 US companies, landed with an unusual result: Anthropic’s Claude held 34.4% of enterprise AI spending, narrowly ahead of OpenAI at 32.3%. A 2.1-point gap, but the first time in either company’s history that Anthropic has led.
What the Billing Data Shows
Ramp’s methodology is straightforward: actual corporate card and reimbursement data, no surveys, no weighting. When 50,000 companies’ purchasing decisions aggregate into a percentage, the number is harder to dispute than any benchmark.
The one-year trajectory is stark. Anthropic quadrupled its enterprise AI adoption rate. OpenAI’s grew by 0.3%. More striking still, OpenAI shed 13.7 percentage points in a single month, its largest single-month decline on record.
Claude Code Driving the Shift
Growth is not spread evenly across Anthropic’s product lineup. The engine is Claude Code, the company’s agentic coding tool, which has displaced significant portions of workflows that previously ran on OpenAI Codex or GitHub Copilot.
Engineering and IT teams consistently cite reliability on long-context tasks, instruction adherence, and multi-step code review as the reasons for switching. The underlying shift: enterprise AI purchasing criteria are moving from “answer quality” toward “workflow embeddability.” A model that reliably completes a 40-step code review matters more to procurement than one that occasionally scores higher on benchmarks but drifts off task.
Traffic and Revenue Also Moving
The Ramp numbers capture enterprise spending, but the same direction shows up elsewhere.
SimilarWeb tracking shows Claude’s monthly web visits grew from 203 million in January 2026 to 824 million in April, a 306% quarterly jump. ChatGPT’s global web-visit share slid from 76.5% in February 2025 to 54.7%, while Gemini climbed to 27.4%.
On revenue, Anthropic’s annualized run rate reached approximately $30 billion, ahead of OpenAI’s $25 billion. Both companies are pre-profit, though Anthropic closed a $65 billion Series H in late May and is projected to post its first operating profit in Q2 2026.
The Apple WWDC Variable
These numbers are a snapshot from before June 8. At WWDC, Tim Cook announced that iOS 27’s Apple Intelligence will let users choose between ChatGPT, Gemini, and Claude, giving Anthropic a native presence on 2.2 billion Apple devices.
At even 5% adoption among Apple Intelligence users, that would add over 100 million new users to Anthropic’s base. The actual distribution effect won’t appear in Ramp data until the Q3 snapshot, but the distribution scale is too large to dismiss.
A Note on the Data
About 16% of companies in the dataset pay for both Claude and ChatGPT simultaneously, roughly one in three AI-using businesses. Ramp’s chief economist cautioned that the market remains in a portfolio phase, where companies hedge across providers rather than commit, and a 2.1-point gap is not a moat.
What May’s numbers settled is narrower but real: OpenAI’s default status in the enterprise market is no longer automatic. For the first time, the billing ledger challenges it.
If this was useful, subscribe to the newsletter for weekly AI PM insights and GenAI case studies.
Further Reading
- Anthropic Files Confidential IPO at $965B Valuation
- Apple WWDC 2026: Siri Rebuilt, Claude Comes to iPhone
Sources
Related Articles
OpenAI Eyes Major Price Cuts as Claude Code Forces an AI Token War
The Wall Street Journal reported June 11 that OpenAI is weighing significant API token price cuts. The trigger: Anthropic's Claude Code drove explosive growth and the company's first profitable quarter. As AI pricing enters a competitive phase, enterprise buyers are gaining leverage.
GPT-5.6 Sol Launches Under Government Lock: Washington's New Frontier AI Gate
OpenAI's GPT-5.6 Sol launched June 26, restricted to ~20 government-vetted partners only. Sol Ultra scores 91.9% on Terminal-Bench 2.1, but the governance framework matters more than the benchmark.