How Does GenAI Influence Creators? Doctoral Student Jiaming Jiang Takes a Look

Doctoral student Jiaming Jiang studies how generative artificial intelligence (genAI) impacts video creator performance and productivity.

Jiaming Jiang is a second-year doctoral student at the Tepper School of Business at Carnegie Mellon University. Jiang studies how technology influences competitive markets and society in general. She is specifically looking at how generative AI (genAI) may transform content creation for both audiences and creators. 

In her paper, “The Impact of GenAI on Content Creation – Evidence from Music Videos,” Jiang looks at how genAI has emerged as an influence on creative industries. Encompassing technologies such as ChatGPT, DALLE2, and Sora, genAI is thought to lower barriers to content creation. Jiang explores the incentives and effects of genAI adoption on a popular Chinese video platform. Her research investigates how genAI adoption affects creator performance in terms of both video popularity and productivity within a competitive creative market. The study concentrates on the music category, where genAI applications are common, and analyzes genAI adoption's influence on creators' performance and the mechanisms behind it. 

Jiang employs a difference-in-difference approach to assess genAI’s impact on creator performance. The findings indicate that genAI adoption affects creators with varying levels of ability (defined based on creators’ viewership prior to adoption) in distinct ways. Specifically, genAI adoption increases production volume among low-ability creators, yet negatively affects the productivity of high-ability creators. While genAI adoption does not uniformly influence creator popularity, it significantly benefits the most capable creators. Further analysis reveals that genAI videos marginally underperform non-genAI videos in viewership across creators and time, a trend mainly driven by low-ability creators’ work. 

Interestingly, the research uncovers a positive spillover effect of genAI adoption on non-genAI videos for high-ability creators. High-ability creators who use genAI also experience increased viewership of their non-genAI videos after adoption. To further explain these findings, Jiang investigates genAI usage patterns and novelty. The research categorizes genAI usage into five main categories: voice synthesizing, AI song cover, music composition, visual genAI, and other. High-ability creators are more likely to apply genAI in more diverse and complex ways, such as voice synthesizing, which may explain their higher viewership.

The research also reveals contrasting adoption patterns between low- and high-ability creators. Low-ability creators were the most likely to adopt genAI initially, but their adoption rate decreased over time. In contrast, high-ability creators were the least willing to adopt initially, but their adoption rate gradually increased. This adoption pattern aligns with the impact of genAI on creator performance. 

This research provides valuable insights into how genAI affects creative content production. The findings demonstrate that genAI disproportionately affects low-, median-, and high-ability creators in terms of performance and productivity. The study also uncovers the spillover effects of genAI adoption on non-genAI videos and explores the underlying mechanisms. This research not only assists content creators in navigating the evolving landscape of AI-driven content creation but also offers valuable insights for creative platforms and their regulatory ecosystems. The research contributes to the ongoing discussion about genAI’s impact on creative platforms. 

Contrary to prevailing research that suggests genAI levels the playing field between low- and high-ability creators, Jiang's study revealed genAI actually increases performance and productivity disparities among low-, median-, and high-ability creators.

Jiang plans to continue working at the intersection of business and technology and look at the policy implications of new and emerging technologies and how they reshape market dynamics.

###

Jiang, Jiaming and Srinivasan, Kannan and Huang, Yan,The Impact of GenAI on Content Creation –Evidence from Music Videos (August 31, 2024). Available at SSRN: https://ssrn.com/abstract=5180595