OpenAI Co-founder's "Vibe Coding" Analysis of AI Risks in U.S. Labor Market, High-Paying Jobs Show Highest Exposure

DeepFlowTech · 2026-03-16T05:50:52+00:00

OpenAI co-founder Andrej Karpathy analyzed the American labor market and found that white-collar positions such as software development have high AI exposure scores of 9, while manual labor and service industry jobs score only 1 to 2. The data sparked misinterpretation and was subsequently deleted. Despite reports supporting AI's impact, Citadel Securities believes there is limited evidence of actual job displacement and predicts that demand for software engineers will increase in 2026.

DeepFlowTech

2026-03-16 05:50:52

Abstract generation in progress

Deep Tide TechFlow News, March 16 — According to Fortune, OpenAI co-founder Andrej Karpathy used AI to analyze the U.S. labor market, rating the exposure of various occupations to AI on a scale of 0 to 10 based on U.S. Bureau of Labor Statistics data. The results show that white-collar jobs such as software development, data science, financial analysis, law, and writing scored as high as 9, with high-paying jobs earning over $100,000 annually averaging a score of 6.7; meanwhile, manual and service jobs like construction workers, caregivers, and barbers scored only 1 to 2. After the data was released, it drew widespread attention, but due to numerous misinterpretations, Karpathy promptly deleted it. Recently, a report from Anthropic reached similar conclusions; however, Citadel Securities countered that there is limited evidence of AI actually replacing jobs at present, with software engineer demand increasing by 11% year-over-year in 2026.

View Original

This page may contain third-party content, which is provided for information purposes only (not representations/warranties) and should not be considered as an endorsement of its views by Gate, nor as financial or professional advice. See Disclaimer for details.

1 Likes