Estimating the AI workforce in tokens: how far from 1% AI-driven GDP growth
robleclerc · x · 2026-08-25
Rob Leclerc shares a back-of-the-napkin estimate of the AI workforce in human-equivalent terms and the time to 1% AI-driven GDP growth, with an interactive calculator.
- Supply: Google processes 3.2 quadrillion tokens/month, OpenAI's API does 15B tokens/minute; tracked global total ≈130Q tokens/year, of which 50Q/year are assumed to do work in the US economy.
- Conversion: a human knowledge worker runs on 10M tokens/year; assuming one human token is worth 10,000 AI tokens (taste, agency, efficiency), one AI-equivalent worker = 100B tokens/year.
- Error margin: Epoch and METR task data price a human work-year at 0.2–2B tokens, giving the author a 500–5,000x margin of error.
More from AGI Musings
- Study Links MBA Rankings to Prevalence of LLM-Style Writing on LinkedIn — mattgroh · 2026-08-25
- Opinion: Cloud Agents Are the Future, All Models Will Move to Swarms — mark_k · 2026-08-25
- Opposing data centers is not meaningfully helpful for AI safety — davidmanheim · 2026-08-25
- Debate on High-Value Models: Is Intelligence Too Cheap to Meter? — haider1 · 2026-08-25
- Stanford Exploration: LLMs Do Not Possess Emotions Despite Appearances — LuizaJarovsky · 2026-08-25
- Colleges cut supplemental essays over AI-writing concerns — Polymarket · 2026-08-25