GLM-52 897 —
GPT-56SC 873 —
CL-OP5X 865 -0.9%
GROK-46H 865 -0.9%
GEM-37FH 865 -0.9%
GPT-56T 861 —
GLM-5 856 —
MUSE-SPK 841 —
QWEN-38X 824 -2.3%
GPT-6A 820 —
KIMI-K3X 810 -1%
CL-FAB5H 787 -0.9%
CL-OP5H 764 -0.9%
CL-OP46H 742 -0.9%
CL-OP47H 733 -1.1%
GEM-38FH 676 -1%
CL-OP47 585 -0.7%
INKL 531 —
CL-OP46 496 -0.2%
CL-OP48 490 -0.2%
GLM-52 897 —
GPT-56SC 873 —
CL-OP5X 865 -0.9%
GROK-46H 865 -0.9%
GEM-37FH 865 -0.9%
GPT-56T 861 —
GLM-5 856 —
MUSE-SPK 841 —
QWEN-38X 824 -2.3%
GPT-6A 820 —
KIMI-K3X 810 -1%
CL-FAB5H 787 -0.9%
CL-OP5H 764 -0.9%
CL-OP46H 742 -0.9%
CL-OP47H 733 -1.1%
GEM-38FH 676 -1%
CL-OP47 585 -0.7%
INKL 531 —
CL-OP46 496 -0.2%
CL-OP48 490 -0.2%
← Back to feed

Anthropic Leads US Enterprise AI Spend for First Time: 41% to OpenAI's 39.5% on Ramp June Index

Ramp’s monthly AI Index for June 2026 shows Anthropic at 41% of US businesses carrying paid AI subscriptions tracked on the platform, compared to OpenAI at 39.5%. The 1.5-point gap is the first time Anthropic has led the index since Ramp began tracking enterprise AI spend. Anthropic gained 2.5 percentage points month-over-month. OpenAI dipped 0.1 points.

What Ramp Tracks

Ramp is a corporate card and expense management platform used by tens of thousands of US companies. Its AI Index measures how many businesses in that universe are running active paid subscriptions to AI providers — a metric that reflects committed spend rather than trial usage or free-tier activity. The data captures enterprise adoption in the moment of actual budget allocation.

The index does not measure revenue, token volume, or seat count. A business paying $20/month for one seat counts the same as one paying $500,000/year. That means the metric specifically reflects breadth of adoption across the company base, not depth within any single account.

What Is Driving the Shift

Claude Code has become the dominant tool in enterprise software development workflows over the past four months. The coding agent has displaced Cursor as the primary AI coding environment across most tracked startup segments, according to separate market analysis. Enterprise deals — including Anthropic’s $150M Claude Corps placement programme, its Glasswing security partnership network, and its 10 finance agent templates targeting professional services — have extended the company’s reach into verticals where OpenAI does not yet have comparable purpose-built products.

Anthropic’s average revenue per user — at $16.20 per user per month versus OpenAI’s $2.20 — suggests the mix is skewed toward paying enterprise users rather than consumer free-tier adoption. That skew would inflate Anthropic’s position on the Ramp index, which counts businesses not individual users.

OpenAI’s Position

OpenAI’s 39.5% figure reflects a platform that is broader in consumer reach but thinner in enterprise monetisation relative to its user count. ChatGPT’s web traffic share fell 22 points over 15 months by one measure; Gemini now holds 27% of AI assistant traffic; Claude’s traffic is up 306% over the same window. The Ramp data adds a spending-confirmation layer on top of those traffic trends.

OpenAI remains the larger company by revenue — its $13B 2025 audited figure versus Anthropic’s $30B annualised run-rate projection — but the enterprise spend composition appears to be shifting in Anthropic’s direction on a per-business basis.

The June figure represents a single data point in a monthly series. Whether the gap widens, holds, or closes in July will depend primarily on whether OpenAI’s Codex, Managed Agents, and enterprise GPT-5.5 rollouts catch up to Anthropic’s lead in agentic coding and vertical-specific products.