- The Industrial Revolution is often cited as a precedent for explosive AI-driven growth, but this analogy is misleading because 10× faster growth was common in pre-industrial Britain (46% of years) while it is extremely rare in modern frontier economies (0.8% of country-years).
- In pre-IR UK, a 10× acceleration (0.66% per year) was within the range of normal year-to-year variation, whereas today a 10× acceleration (e.g., 19% for the US) would be multiple standard deviations outside historical experience.
- Using a standard deviation approach, an Industrial Revolution-style shift in the mean growth rate would take modern frontier economies to about 2.8% per year, far below the 10× claim often advanced.
- Most modern instances of very high growth (>20%) are driven by oil windfalls or post-conflict recovery, not sustainable technological acceleration, and thus do not provide a meaningful precedent for AI-driven growth.
- The author argues that the IR precedent actually supports skepticism about extremely rapid growth, because the required acceleration today is qualitatively different from what occurred in the 1700s.