Occupational, industry, and geographic exposure to artificial intelligence: A novel dataset and its potential uses.

100%

Occupational, industry, and geographic exposure to artificial intelligence: A novel dataset and its potential uses.

Abstract

Research Summary: We create and validate a new measure of an occupation's exposure to AI that we call the AI Occupational Exposure. We use the AI Occupational Exposure to construct a measure of AI exposure at the industry level, which we call the AI Industry Exposure, and a measure of AI exposure at the county level, which we call the AI Geographic Exposure. We also describe several ways in which the AI Occupational Exposure can be used to create firm level measures of AI exposure. We validate the measures and describe how they can be used in different applications by management, organization and strategy scholars.

Managerial Summary: Although artificial intelligence promises to spur economic growth, there is widespread concern that it could displace workers, alter industry trajectories, and reshape organizations. Despite the interest in this area, we have limited ability to study the effects of AI on occupations, firms, industries, and geographies because of limited availability of data that measures exposure to AI. To address this limitation, we create and validate a new measure of an occupation's exposure to AI that we call the AI Occupational Exposure. We use the AI Occupational Exposure to construct a measure of AI exposure at the industry level.

and county level. We describe how our measures can be useful to scholars and policy-makers interested in identifying the effect of AI on markets.

One INTRODUCTION

Recent advances in artificial intelligence have generated excitement about AI's potential to spur economic growth, and scholars believe that AI has the potential to be "the most important general-purpose technology of our era." However, there is concern that advances in AI may also have significant consequences for labor markets, firms, and industries by displacing workers, transforming occupational jurisdictions, altering strategy, and affecting performance. For decades now, scholars have considered whether and how rapid advances in information technologies change the nature of competition and strategy. In recent years, researchers have increasingly begun to investigate how AI affects firm design, strategy, organizational learning, and management. However, despite great interest in the academic literature and public press about AI's effects on occupations, firms, and markets, there has been little systematic collection of evidence. Part of the reason for the lack of evidence is that the rapid advancement in AI is a nascent phenomenon, and accordingly, appropriate tools to measure its impact have yet to be developed.

In order to fill this gap, we develop a new measure of the exposure to AI across occupations that we call the AI Occupational Exposure. This measure links common and general applications of AI to workplace abilities and occupations. We show the potential of the occupation-level AI Occupational Exposure by using it to construct two derivative measures. We aggregate the AI Occupational Exposure to the industry level to construct an AI Industry Exposure as well as to the county-level to construct a geographic measure of AI exposure that we call the AI Geographic Exposure. We also describe several ways in which the AI Occupational Exposure can be used to create firm level measures of AI exposure. We validate our AI Occupational Exposure measure and then describe how these measures can be used in different applications by strategy, management, and innovation scholars. We have made these datasets freely available for use by scholars, policy-makers, and practitioners.

We build our measures by linking common AI applications to occupational abilities using a crowd-sourced dataset. We aggregate the effect at the ability level to construct a measure that identifies the potential exposure of occupations to AI. While similar in spirit to recent work, our methodology is unique in that it links specific applications of AI to occupational abilities and is agnostic as to whether AI substitutes for or complements workplace abilities and occupations. We discuss differences with existing datasets in more detail below. We believe our methodology has the potential to be used in a variety of applications within the field of strategy and, as we describe below, can be used to construct AI exposure at the firm level as well.

Our article contributes in several ways. First, we develop a new methodology linking applications of AI technology to human abilities. We use this method to generate occupation-level, industry-level, and geographic-level AI Exposure measures, which are available for other researchers to use. By describing the data construction and validation in the body of this article and providing this data for use by other researchers, we are answering a call for more articles that develop and describe datasets for other researchers to use. Second, we situate our measures within existing data by comparing the AI Occupational Exposure with other measures that link work and occupations to AI, automation, and robotics. Third, we describe several ways in which the measures can be used by researchers to study the effects of exposure to AI on workers, firms, industries, and geographies.

The article proceeds as follows. In the next section, we outline the methodology that we use to construct the AI Occupational Exposure and its accompanying AI Industry Exposure and AI Geographic Exposure measures. In the third section, we describe the steps we take to validate our methodology and resulting measures. In the fourth section, we describe potential applications for the datasets, with particular attention to the use of the AI Occupational Exposure for organizational level studies. The fifth section concludes.

Two CONSTRUCTION OF THE AI EXPOSURE MEASURES

AI. Two point one. Data

Two point two. Methodology

Two point two point two. Linking AI applications to O*NET abilities

Two point two point three. Calculating occupational exposure

Two point two point four AIIE and AIGE

Three VALIDATION OF THE AI OCCUPATION EXPOSURE SCORES

Three point one Most- and least-exposed occupations and industries

Three point two. Discussion of selected occupations

Three point two point one. Surgeons and slaughterers

Three point two point two. Mathematical technicians and accountants and auditors

Three point three. Quantitative validation

Four. Potential applications of AIOE, AIIE, and AIGE

Four point one. Distinctions between the AIOE and existing datasets

Four point two | Potential uses of the AIOE, AIIE, and AIGE across fields

Four point two point two Human capital |

Four point two point three Organizational design

Four point two point four Industry evolution

Four point two point five Agglomeration and location-based advantages

Five | CONCLUSION

SUPPORTING INFORMATION

Overview

The document presents a new methodology for measuring occupational, industry, and geographic exposure to artificial intelligence, addressing the limitations in current data availability. The proposed AI Occupational Exposure dataset aims to serve scholars and policymakers by highlighting AI's potential impacts on workplaces and industries.

Key Points

1The AI Occupational Exposure dataset measures exposure across occupations, industries, and geographies
2A new methodology links specific AI applications to occupational abilities
3The dataset is designed for use by management, strategy scholars, and policymakers
4Validation of the measures allows for accurate assessment of AI's impact on the workforce
5Datasets are made freely available to enhance research on AI exposure effects.

Details

Authors: Edward Felten, Manav Raj, Robert Seamans
Category: Technology and Engineering

PDF
Proficiency of records practitioners using artificial intelligence to enhance records management practices at the Gauteng Department of Education in South Africa
This study examines the proficiency of records practitioners at the Gauteng Department of Education in using artificial intelligence (AI) to improve records management practices. It employs a mixed-methods research approach to evaluate the existing skill levels and proposes a framework for upskilling these practitioners in AI utilization.
PDF
Intelligent Archive Management Based on Deep Learning Technology Driven by Artificial Intelligence
This research article presents a study on an intelligent archive management system (AMS) that employs deep learning (DL) technology to enhance the efficiency and accuracy of processing multimodal data. The proposed model addresses traditional AMS shortcomings such as data classification accuracy and system responsiveness.
PDF
Discharging Records Management Activities Using Artificial Intelligence at the Council for Scientific and Industrial Research, South Africa
This study investigates the application of artificial intelligence (AI) in enhancing records management activities at the Council for Scientific and Industrial Research (CSIR) in South Africa. It provides a framework for utilizing AI to improve efficiency in managing records throughout their life cycle.
PDF
Human-like systematic generalization through a meta-learning neural network
This article explores the capacity of neural networks to achieve human-like systematicity through a meta-learning approach that enhances compositional skills. It presents experimental findings comparing human and machine performance on systematic generalization tasks.
PDF
TBI-603 Network Security & Cyber Law Unit Notes
This document serves as comprehensive unit notes for a course on Network Security and Cyber Law, covering topics like network security practices, types of security attacks, and legal aspects of cybersecurity.