As AI's capability grows, what once-human tasks will it be able to do, by when? What will those displaced humans do? We look to the automation revolutions of the past to see what the historical pattern has been, and explore in what ways AI is different that could change the pattern.
The pattern will change. The implications are immense.
Sources
Leontief's horse. Wassily Leontief, 1983, National Academy of Engineering symposium The Long-Term Impact of Technology on Employment and Unemployment. Quote and horse-population figures via Brynjolfsson & McAfee, "Will Humans Go the Way of Horses?", Foreign Affairs (2015): https://www.foreignaffairs.com/world/will-humans-go-way-horses
Software developer pay. U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, Software Developers (SOC 15-1252): https://www.bls.gov/oes/2021/may/oes151252.htm · https://www.bls.gov/oes/2022/may/oes151252.htm · https://www.bls.gov/oes/2023/may/oes151252.htm · Occupational Outlook Handbook: https://www.bls.gov/ooh/computer-and-information-technology/software-developers.htm · Total-comp figure: Levels.fyi (2026).
AI timelines (experts). Katja Grace et al., "Thousands of AI Authors on the Future of AI" (2023 survey, 2,778 researchers): https://arxiv.org/abs/2401.02843 · https://aiimpacts.org/wp-content/uploads/2023/04/Thousands_of_AI_authors_on_the_future_of_AI.pdf
AI timelines (forecasters). Metaculus (community forecasts; live figures): "first general AI system" https://www.metaculus.com/questions/5121/ · "weakly general AI" https://www.metaculus.com/questions/3479/
Goldman Sachs. "An AI Job Apocalypse?", Goldman Sachs Research, June 25, 2026: https://www.goldmansachs.com/insights/top-of-mind/an-ai-job-apocalypse (Goldman's own view is that the disruption is temporary.)
Occupational exposure. Tyna Eloundou, Sam Manning, Pamela Mishkin, Daniel Rock, "GPTs are GPTs," Science 384 (2024): https://www.science.org/doi/10.1126/science.adj0998 · working paper: https://arxiv.org/abs/2303.10130
Entry-level cracks. Stanford Digital Economy Lab, "Canaries in the Coal Mine? Six Facts about the Recent Employment Effects of AI" (Nov 2025): https://digitaleconomy.stanford.edu/app/uploads/2025/11/CanariesintheCoalMine_Nov25.pdf
Current labor data. Maxim Massenkoff & Peter McCrory, "Labor Market Impacts of AI: A New Measure and Early Evidence," Anthropic (Mar 5, 2026): https://www.anthropic.com/research/labor-market-impacts (Note: Anthropic funds both the models and this research.)
Depression unemployment anchor. U.S. unemployment peaked near 25% in 1933 (verify exact figure before citing on air).
Hosted on Acast. See acast.com/privacy for more information.
