Experimental evidence on the productivity effects of generative artificial intelligence

Abstract

We examined the productivity effects of a generative artificial intelligence (AI) technology, the assistive chatbot ChatGPT, in the context of midlevel professional writing tasks. In a preregistered online experiment, we assigned occupation-specific, incentivized writing tasks to 453 college-educated professionals and randomly exposed half of them to ChatGPT. Our results show that ChatGPT substantially raised productivity: The average time taken decreased by 40% and output quality rose by 18%. Inequality between workers decreased, and concern and excitement about AI temporarily rose. Workers exposed to ChatGPT during the experiment were 2 times as likely to report using it in their real job 2 weeks after the experiment and 1.6 times as likely 2 months after the experiment.

Unleashing efficiency and insights: Exploring the potential applications and challenges of ChatGPT in accounting

“…Eloundou et al (2023) list accountants, auditors, and tax preparers as having 100% exposure to significant automation. Noy and Zhang (2023) demonstrate that ChatGPT leads to enhanced productivity, job satisfaction, and self-efficacy, with a notable advantage for low-ability workers. Although there are papers discussing ChatGPT's implementation in different disciplines such as education, healthcare, medicine, environment, and law (e.g., Aydın & Karaarslan, 2022;Choi et al, 2023;Frye, 2022;Mann, 2023), rare scholars have explored the promising applications of ChatGPT in the accounting field (De Kok, 2023;Gu et al, 2023;Noy & Zhang, 2023).…”

“…Noy and Zhang (2023) demonstrate that ChatGPT leads to enhanced productivity, job satisfaction, and self-efficacy, with a notable advantage for low-ability workers. Although there are papers discussing ChatGPT's implementation in different disciplines such as education, healthcare, medicine, environment, and law (e.g., Aydın & Karaarslan, 2022;Choi et al, 2023;Frye, 2022;Mann, 2023), rare scholars have explored the promising applications of ChatGPT in the accounting field (De Kok, 2023;Gu et al, 2023;Noy & Zhang, 2023). In the rapidly evolving AI landscape, our study serves as one of the initial explorations of ChatGPT in accounting, echoing the calls made in accounting literature for more research dedicated to emerging technologies within the accounting domain (Fisher et al, 2016;Sutton et al, 2016).…”

“…As the latest development in AI and NLP technology exploration, ChatGPT should have attracted widespread attention from scholars. However, accounting research related to ChatGPT is rather limited up to now (De Kok, 2023;Gu et al, 2023;Noy & Zhang, 2023). Our study serves as an initial exploration of ChatGPT's facilitation in accounting and contributes to the growing body of knowledge in this rapidly iterative AI field.…”

The now and future of ChatGPT and GPT in psychiatry

“…In research, ChatGPT is renowned as an expert writing assistant. A recent experiment showed that college‐educated professionals using ChatGPT enjoyed substantially increased productivity in occupation‐specific, incentivized writing tasks 28 . Using ChatGPT to polish later drafts of academic writing for improved readability and language is recommended.…”

In Silico Design of Freeform Solar Cell Structures from High‐Throughput Artificial Intelligence‐Generated Configurations

“…Unlike traditional automation processes, which rely on explicitly programmed sequences, ML is designed to handle large datasets and high variability across both predictive and generative tasks. [1] The success of models such as ChatGPT in natural language processing (NLP) [2] and DALL-E in computer vision (CV) [3] has drawn public attention to the emerging field of AIGC. Beyond NLP and CV, AIGC is becoming a crucial scientific tool for addressing inverse design challenges, aimed at discovering new materials and structures with targeted functionalities.…”