Dynamic Personality-Aware Persuasion: Adapting Messaging Strategies in Real Time Using AI Feedback Loops

Building on Lou & Xu’s (2025) agent-based modeling of personality and persuasion, and the work by Matz et al. (2024) on LLM-powered personalization, this project proposes a real-time, adaptive persuasion framework. Instead of treating personality as a fixed trait, the system would use continuous behavioral data and user feedback to update its model of the user’s current mood, openness, and receptiveness. The AI would then adjust its persuasive strategy accordingly—perhaps becoming more evidence-based or emotionally supportive as the situation demands. This approach challenges the static assumption of most personality-targeted persuasion literature and leverages cutting-edge AI techniques (Meguellati, 2025) to create a “living persuasion model.” The novelty lies in its closed-loop, adaptive nature, promising more authentic and contextually appropriate interactions, and has high potential for applications in customer service bots, health interventions, and misinformation correction.

References:

1. Personality Modeling for Persuasion of Misinformation using AI Agent. Qianmin Lou, Wentao Xu (2025). arXiv.org.
2. The potential of generative AI for personalized persuasion at scale. S. C. Matz, J. D. Teeny, S. Vaid, H. Peters, G. M. Harari, M. Cerf (2024). Scientific Reports.
3. The Duality of Persuasion: Between Personalization and Detection. Elyas Meguellati (2025). The Web Conference.