Generative artificial intelligence usage by researchers at work: Effects of gender, career stage, type of workplace, and perceived barriers

TL;DR

This study investigates how factors like gender, career stage, workplace type, and perceived barriers influence researchers' use of generative AI, revealing significant disparities and insights into adoption patterns across different groups.

Contribution

It employs a regression model to isolate the effects of various factors on AI usage, controlling for potential biases and providing new insights into adoption dynamics.

Findings

Researchers facing barriers show 11% increased AI use

Female researchers use AI 7% less than males

Advanced career researchers use AI 19% less

Abstract

The integration of generative artificial intelligence technology into research environments has become increasingly common in recent years, representing a significant shift in the way researchers approach their work. This paper seeks to explore the factors underlying the frequency of use of generative AI amongst researchers in their professional environments. As survey data may be influenced by a bias towards scientists interested in AI, potentially skewing the results towards the perspectives of these researchers, this study uses a regression model to isolate the impact of specific factors such as gender, career stage, type of workplace, and perceived barriers to using AI technology on the frequency of use of generative AI. It also controls for other relevant variables such as direct involvement in AI research or development, collaboration with AI companies, geographic location, and scientific discipline. Our results show that researchers who face barriers to AI adoption experience an 11% increase in tool use, while those who cite insufficient training resources experience an 8% decrease. Female researchers experience a 7% decrease in AI tool usage compared to men, while advanced career researchers experience a significant 19% decrease. Researchers associated with government advisory groups are 45% more likely to use AI tools frequently than those in government roles. Researchers in for-profit companies show an increase of 19%, while those in medical research institutions and hospitals show an increase of 16% and 15%, respectively. This paper contributes to a deeper understanding of the mechanisms driving the use of generative AI tools amongst researchers, with valuable implications for both academia and industry.