The Reality of AI Spending Tiers

Recent data from the Ramp AI Index reveals a massive disparity in how companies allocate budget toward artificial intelligence. While industry headlines suggest that compute costs are beginning to rival payroll, the data indicates this is currently a phenomenon limited to the top 1% of adopters, whom the index labels as "AI-pilled."

  • The Top 1%: These firms spend $7,500 per employee per month on AI. For context, this remains below the average monthly salary of a software engineer (approximately $16,000), suggesting that while AI costs are significant, they have not yet eclipsed human capital for most organizations.
  • The Top 10%: Spending drops significantly to $611 per employee per month.
  • The Median: The typical firm spends only $11.38 per employee per month, which largely reflects the cost of standard enterprise software seat licenses.

Despite the wide variance in spending, the trajectory for high-intensity users is upward. AI-pilled firms saw a 14.1% increase in per-employee AI spending over the last month. To manage these escalating costs, these power users are increasingly adopting a "mix and match" strategy. Rather than relying on a single vendor, they are actively rotating between multiple frontier models and integrating cheaper, open-source alternatives to optimize their token consumption and maintain operational efficiency.