AI at Work

Ramp data shows top 1% of businesses drive 80% of AI spend

Ramp data reveals that a tiny fraction of VC-backed startups dominates spending on OpenAI and Anthropic services. This extreme concentration raises questions about the sustainability of current AI revenue models as broader enterprise adoption begins to accelerate.

· 1 source · AI spending concentration

Ramp data indicates that the top 1% of businesses account for 80% of spending volume for OpenAI and Anthropic through its platform. Within this elite group, 80% of customers are venture capital-backed startups, with much of the token volume used for internal purposes or wrapped in services featuring VC-subsidized discounts. This pattern mirrors broader trends in the tech sector; Nvidia reported that two customers made up 39% of its second-quarter revenue, while Anthropic's IPO prospectus noted that a fifth of its first-half revenue came from its top two customers.

Despite this heavy reliance on a small cohort of high-spending startups, signs of broader adoption are emerging. Businesses in the 10th percentile of spending grew from around $200 per employee to around $800 per employee since summer, suggesting an expanding adoption curve. The key risk remains whether this concentration will persist as buyers mature or if it is a temporary phase of the adoption cycle. If venture capital funding dries up, the current revenue structure could face significant pressure. Once Anthropic and OpenAI go public, their audited financials will reveal the true quality and concentration of their revenue, providing a clearer picture of the market's underlying health.