🤖 AI Summary
This study empirically investigates the diffusion patterns and community response mechanisms of AI-generated content (AIGC) in Reddit’s art-related subreddits. Using large-scale temporal API data collected prior to 2023, we integrate text classification (to identify AI-self-labeled posts and skeptical comments), community rule coding, and sentiment trend analysis. Our findings reveal: (1) AI-generated images constitute a negligible share (<0.5%) of art submissions and exhibit a sustained decline, yet user skepticism intensifies—manifesting in more frequent critiques and increasingly negative sentiment; (2) AIGC boosts newcomer posting rates, but only fosters positive engagement in communities with explicit, well-enforced rules; (3) critically, we demonstrate for the first time that regulatory ambiguity exacerbates negative discourse, whereas transparent, codified policies mitigate conflict. These results provide rigorous empirical evidence to inform platform governance strategies and advance understanding of AIGC’s socio-technical impact on online creative communities.
📝 Abstract
Broadly accessible generative AI models like Dall-E have made it possible for anyone to create compelling visual art. In online communities, the introduction of AI-generated content (AIGC) may impact social dynamics, for example causing changes in who is posting content, or shifting the norms or the discussions around the posted content if posts are suspected of being generated by AI. We take steps towards examining the potential impact of AIGC on art-related communities on Reddit. We distinguish between communities that disallow AI content and those without such a direct policy. We look at image-based posts in these communities where the author transparently shares that the image was created by AI, and at comments in these communities that suspect or accuse authors of using generative AI. We find that AI posts (and accusations) have played a surprisingly small part in these communities through the end of 2023, accounting for fewer than 0.5% of the image-based posts. However, even as the absolute number of author-labeled AI posts dwindles over time, accusations of AI use remain more persistent. We show that AI content is more readily used by newcomers and may help increase participation if it aligns with community rules. However, the tone of comments suspecting AI use by others has become more negative over time, especially in communities that do not have explicit rules about AI. Overall, the results show the changing norms and interactions around AIGC in online communities designated for creativity.