Welfare Implications of Democratization in Content Creation
Preprint
- 1 January 2023
- preprint
- Published by Elsevier in SSRN Electronic Journal
Abstract
The rise of content-creation technologies such as generative artificial intelligence (G-AI) has reshaped the way people produce and consume content. These technologies democratize content creation, enabling a broader array of inexperienced individuals to generate content that rivals the quality produced by professionals. In this paper, we adopt a game-theoretic model to study how the democratization of content creation affects consumer welfare. While the content-democratization technology can assist low-quality creators in improving their content quality or creation efficiency, it can precipitate a flood of low-quality content in the market and crowd out high-quality content, leading to a loss in consumer welfare. The welfare loss occurs when the content quality or creation efficiency of the low-quality creators is at a low level. We further show that it is more likely to observe such a welfare loss when the accessibility of the technology shrinks or when content platforms can screen content and push more high-quality content to consumers. Moreover, we investigate the welfare implications when the technology simultaneously uplifts the quality of all content. In addition, we find that if the content-democratization technology can learn from its usage or human-generated content in the past, consumer welfare may decline. Furthermore, we explore a specific content type (e.g., selfie posts on Instagram) for which high-quality content can yield a disutility for consumers.Keywords
This publication has 26 references indexed in Scilit:
- Predictive Analytics and Ship-Then-Shop SubscriptionManagement Science, 2024
- Survey of Hallucination in Natural Language GenerationACM Computing Surveys, 2023
- Competing for Recommendations: The Strategic Impact of Personalized Product Recommendations in Online MarketplacesMarketing Science, 2023
- Implications of Revenue Models and Technology for Content Moderation StrategiesMarketing Science, 2022
- Collusion by Algorithm: Does Better Demand Prediction Facilitate Coordination Between Sellers?Management Science, 2019
- Competing for Attention in Social Communication MarketsManagement Science, 2016
- Differentiation with User-Generated ContentManagement Science, 2015
- User-Generated Content and Bias in News MediaManagement Science, 2013
- Information congestionThe RAND Journal of Economics, 2009
- Information Overload in a Network of Targeted CommunicationThe RAND Journal of Economics, 2004