Economist Tom Davidson’s 2023 “compute-centric” model answers two questions: when could AI automate all current human labor, and how fast might that transition happen?
Davidson's model considers the resources required to train increasingly capable AI systems (compute × algorithmic efficiency) and the rate at which these resources grow. Simulating the model suggests a median arrival of AI that can automate 100% of human labor in 2043, with the jump from 20% to 100% automation taking roughly three years.
Dive deeper or test alternative assumptions for yourself:
• Interactive website — https://takeoffspeeds.com
• Tom Davidson's report — https://www.openphilanthropy.org/research/what-a-compute-centric-framework-says-about-takeoff-speeds/
Nathan Metzger
2025-06-02 04:11:04 +0000 UTC