When
Where
Kelli Questrom Associate Professor of Information Systems Management & Computing and Data Sciences, Questrom School of Business, Boston University
Generative AI, Human Creativity, and Art
Abstract: Recent artificial intelligence (AI) tools have demonstrated their ability to produce outputs traditionally considered creative. One such system is text-to-image generative AI (e.g., Midjourney, Stable Diffusion, Dall-E), which automates humans’ execution to generate high-quality digital artworks. Utilizing a dataset of over 4 million artworks from more than 50,000 unique users, our research shows that text-to-image AI substantially enhances human creative productivity by 25% and increases the value as measured by the likelihood of receiving a favorite per view by 50% over time. While peak artwork content novelty (focal objects and object relationships) increases over time, average content novelty declines, suggesting an expanding but inefficient creative space. Additionally, there is a consistent reduction in both peak and average visual novelty (pixel-level stylistic elements). Importantly, AI-assisted artists who can produce more novel content ideas, regardless of overall novelty before adoption, produce artworks that their peers evaluate more favorably. The results imply that ideation and likely filtering are necessary skills in the text-to-image process, thus giving rise to “generative synesthesia” - the harmonious blending of human senses and AI mechanics to discover new creative workflow. Lastly, AI adoption decreased value capture (favorites earned) concentration among the adopted.
Bio: Dokyun Lee studies the responsible application, development, and impact of AI in digital consumer and e-commerce analytics with a focus on text data. Specific interests are Generative AI, Economics of unstructured data (content extraction, understanding, engineering, marketing), and Unintended Consequence of AI in Business in the context of digital consumer management, platform design, market competition, advertising, human-ai collaboration, innovation, and creativity. Dokyun run the BIT (Business Insights through Text) Lab. His work has been published at journals and conferences including Management Science, Information Systems Research, Journal of Marketing Research, AAAI, AIES, and WWW. Dokyun is a recipient of ISS Gordon B Davis Young Scholar, 2021 Marketing Science Institute Young Scholar, ISS Sandra A. Slaughter Early Career Award, CDO Magazine Academic Data Leader, ISR Best Paper Award, MSBA Teaching Award, Management Science Distinguished Service Award, and Best Conference Paper Awards. His work is supported by organizations such as Adobe, Bosch Institute, Google Cloud, Marketing Science Institute, McKinsey & Co, Nvidia, Net Institute, and Prudential Foundation. He holds a Bachelor’s degree [2009] in Computer Science from Columbia University (Machine Learning Focus), a Master’s degree [2010] in Statistics (Master's Thesis: Johnson-Lindenstrauss Lemma and its Effect on Supervised Learning) from Yale University and PhD [2015] from the Operation, Information and Decisions department of the Wharton School (Thesis: Three Essays in Big Data Consumer Analytics in E-Commerce). Before academia, Dokyun worked at 4 tech start-ups and Blackrock as a quantitative software engineer and at Thomson Reuters as an ML contractor building a natural language processing engine for financial data.