# AI Spend Reprices to Payroll Scale

Enterprise AI budgets are migrating from per-seat SaaS licensing to consumption pricing benchmarked against salaries, and falling token prices expand usage faster than they cut cost, so aggregate AI spend keeps rising.

- Conviction: 40 / 100 (forming)
- Horizon: Emerging (watchlist)
- Tracking since: 2026-07-17T00:00:00.000Z
- Last updated: 2026-07-17T05:34:02.889Z
- Canonical: https://polylog.news/ai/trends/ai-token-spend-payroll-scale
- Publisher: Polylog
- Affected regions: Global

## Recent evidence

- [confirms] At the AI Frontier, Token Bills Are Approaching Payroll Scale, a16z and Ramp Find (2026-07-17): a16z and Ramp data cited in the piece put the heaviest corporate AI users at ~$90,000 per employee a year on model tokens, with cheaper tokens accelerating rather than capping spend — a direct instance of consumption pricing benchmarked to salaries and the Jevons dynamic the thesis predicts.