The Potential Impact of Disruptive AI Innovations on U.S. Occupations

2025-07-15

AI Summary

This study systematically examines the structural impact of AI innovation on U.S. occupational markets, distinguishing between “consolidating” AI—reinforcing existing task divisions—and “disruptive” AI—reconfiguring task boundaries. Leveraging 3,237 U.S. AI patents, we construct an original “AI Disruption Index” to quantitatively identify and differentiate these two innovation types for the first time. By mapping patent-level technical features to O*NET occupational task data, we conduct large-scale, region- and industry-level causal inference analyses. Results show that consolidating AI follows conventional automation logic, primarily substituting routine physical and cognitive tasks; in contrast, disruptive AI significantly reshapes high-order cognitive work, intensifying labor market tensions—particularly in coastal tech hubs and skill-scarce regions. Our index provides a novel, empirically grounded measurement tool for AI governance and regional workforce policy design.

Technology Category

• Philosophy and Ethics of AI: AI & Jobs/Labor
• Humans and AI: Other Foundations of Human Computation & AI
• Computer Vision: Remote Sensing / Geospatial AI

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

The rapid rise of AI is poised to disrupt the labor market. However, AI is not a monolith; its impact depends on both the nature of the innovation and the jobs it affects. While computational approaches are emerging, there is no consensus on how to systematically measure an innovation's disruptive potential. Here, we calculate the disruption index of 3,237 U.S. AI patents (2015-2022) and link them to job tasks to distinguish between "consolidating" AI innovations that reinforce existing structures and "disruptive" AI innovations that alter them. Our analysis reveals that consolidating AI primarily targets physical, routine, and solo tasks, common in manufacturing and construction in the Midwest and central states. By contrast, disruptive AI affects unpredictable and mental tasks, particularly in coastal science and technology sectors. Surprisingly, we also find that disruptive AI disproportionately affects areas already facing skilled labor shortages, suggesting disruptive AI technologies may accelerate change where workers are scarce rather than replacing a surplus. Ultimately, consolidating AI appears to extend current automation trends, while disruptive AI is set to transform complex mental work, with a notable exception for collaborative tasks.

Problem

• Measuring disruptive potential of AI innovations on jobs
• Distinguishing consolidating vs disruptive AI impacts on tasks
• Analyzing regional and sectoral disparities in AI labor effects

Innovation

• Calculate disruption index of AI patents
• Link patents to job tasks
• Distinguish consolidating and disruptive AI