The top 1% of U.S. companies now spend $7,500 per employee per month on AI services—less than half the $16,000 monthly cost of an average software engineer, according to the Ramp AI Index released this week. The data, drawn from transaction records across thousands of businesses, shows that even aggressive adopters remain far below human salary costs.

Ramp segments firms into three tiers based on monthly per-employee AI spend. The gap between them is massive:

• The top 1% averages $7,500 per employee.
• The top 10% spend $611 per employee.
• The median firm allocates $11.38 per employee—roughly one enterprise SaaS seat.

Growth accelerated in May. The highest-spending tier increased per-employee costs by 14.1% compared to April. These firms route workloads across multiple frontier models and open-source platforms, balancing performance against cost. Mercor's CEO stated the startup spends more on tokens for internal agents than on employee headcount. An Nvidia executive noted that compute expenses now exceed employee salaries at the company.

For software architects and budget managers, the $7,500 figure serves as a planning reference for heavy integration. Expect monthly costs between $600 and $1,000 per employee when mixing hosted and self-hosted models. The current gap suggests most firms face a decision: remain a median adopter or shift toward the high-utility spending patterns of the top decile. Read more: Snowflake signs $6B AWS chip deal—why your cloud costs will drop this fall.