Paper Type
Complete
Abstract
Online platforms often experience attention shocks where viral narratives rapidly dominate discussions. This study examines how such a shock shaped creator behavior on Bilibili during the late-2025 diffusion of the “kill line” discourse. We focus on a creator-side dilemma: whether responding to a rapidly spreading but misleading narrative improves audience growth. We analyze a panel of 16,208 creators (Sep 2025–Jan 2026) using a hybrid LLM-RoBERTa pipeline to identify misinformation at scale. Employing Propensity Score Matching and Difference-in-Differences, we find that the shock increased misinformation production by 7.6% but yielded no significant aggregate effect on follower growth. Crucially, subgroup analysis reveals a stark asymmetry: smaller creators experienced an 81.0% increase in weekly follower growth, while established creators faced a 33.0% decline. These findings suggest that the returns to misinformation are structurally contingent, challenging the assumption that sensationalism uniformly rewards all producers in digital ecosystems.
Paper Number
1929
Recommended Citation
Sun, Yuanze and Zhang, Xiaohui, "Asymmetric Returns to Misinformation: The “Kill Line” Case" (2026). AMCIS 2026 Proceedings. 27.
https://aisel.aisnet.org/amcis2026/sigadit/sigadit/27
Asymmetric Returns to Misinformation: The “Kill Line” Case
Online platforms often experience attention shocks where viral narratives rapidly dominate discussions. This study examines how such a shock shaped creator behavior on Bilibili during the late-2025 diffusion of the “kill line” discourse. We focus on a creator-side dilemma: whether responding to a rapidly spreading but misleading narrative improves audience growth. We analyze a panel of 16,208 creators (Sep 2025–Jan 2026) using a hybrid LLM-RoBERTa pipeline to identify misinformation at scale. Employing Propensity Score Matching and Difference-in-Differences, we find that the shock increased misinformation production by 7.6% but yielded no significant aggregate effect on follower growth. Crucially, subgroup analysis reveals a stark asymmetry: smaller creators experienced an 81.0% increase in weekly follower growth, while established creators faced a 33.0% decline. These findings suggest that the returns to misinformation are structurally contingent, challenging the assumption that sensationalism uniformly rewards all producers in digital ecosystems.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
SIG ADIT