Reproduce “AI Persuasive Framing in Collective Dilemmas” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Anders Giovanni Møller, Alessia Galdeman, Arianna Pera, Luca Maria Aiello
An investigation into AI assistants employing personalized persuasive framing in an iterated Collective Risk Game found that while initial prosocial nudges boosted cooperation, antisocial persuasion led to more substantial and persistent reductions in collective contributions and group success.
Key insights
- Personalized persuasive AI interventions can initially enhance cooperation in collective dilemmas, demonstrating a stronger impact than simpler informational or static message framing.
- A significant asymmetry exists in the persistence of AI's persuasive effects: antisocial persuasion promoting selfishness is more robust and enduring than prosocial persuasion encouraging cooperation.
- The findings highlight the dual-use potential of AI, indicating that systems designed to influence group behavior could more persistently undermine collective action if misaligned or maliciously employed.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.