Countering Online Problematic Speech Through “More Speech:” How Do Different Counterspeech Strategies and AI-Powered Interventions Shape Community Dynamics?
Access to this document is restricted. Some items have been embargoed at the request of the author, but will be made publicly available after the "No Access Until" date.
During the embargo period, you may request access to the item by clicking the link to the restricted file(s) and completing the request form. If we have contact information for a Cornell author, we will contact the author and request permission to provide access. If we do not have contact information for a Cornell author, or the author denies or does not respond to our inquiry, we will not be able to provide access. For more information, review our policies for restricted content.
Problematic speech pervades social media, causing emotional distress and eroding democratic dialogue. Counterspeech offers a promising solution that balances mitigating online harm with safeguarding freedom of expression. However, despite audience reactions playing a crucial role in shaping and reinforcing community norms, little is known about how audiences respond to different types of counterspeech on social media. To fill this critical gap, this dissertation presents three experimental studies, conducted on a simulated social media platform/webpage, to examine which types of counterspeech, delivered by whom, and under what conditions can effectively garner audience endorsement and explore how to scale up socially endorsed counterspeech to facilitate prosocial community norms. Study 1 investigates the audience’s emotional, moral, and attitudinal responses to different counterspeech strategies, revealing that audiences consistently prefer morality-based counterspeech even when responding to severe offense and that moral judgment is the primary driver behind audience endorsement. Extending these findings through the lens of justice theory, Study 2 examines actual audience behaviors (upvotes/downvotes) and community perceptions toward justice-oriented counterspeech. Results show that restorative justice-based counterspeech effectively serves justice, attracts greater audience approval, and enhances community satisfaction, particularly when offenders are viewed as morally corrigible. Building upon the socially endorsed counterspeech strategies identified in Studies 1 and 2, Study 3 evaluates audience acceptance of AI-delivered morality- and empathy-based counterspeech. Although human speakers retain greater moral authority than AI agents, audiences perceive AI-generated counterspeech as moderately permissible and authentic, suggesting a promising pathway for using generative AI to scale socially endorsed counterspeech interventions. Taken together, this dissertation provides a novel framework for understanding moral dynamics underlying digital discourse. By revealing the effective counterspeech in drawing audience endorsement and demonstrating the potential of AI-powered counterspeech, this work offers a roadmap for developing theoretically grounded, practically scalable counterspeech interventions, contributing to fostering healthier and more constructive online communities.