Problem
Research questions and friction points this paper is trying to address.
persuasion
opinion change
human-AI interaction
longitudinal dataset
Large Language Models
Talk2AI: A Longitudinal Dataset of Human--AI Persuasive Conversations
🤖 AI Summary
This study addresses the scarcity of large-scale longitudinal datasets capturing human–AI dialogues that support research on persuasion, attitude change, and human–AI interaction. We present a novel dataset comprising 3,080 multi-turn conversations (30,800 utterances) between 770 Italian adults and four leading large language models—GPT-4o, Claude Sonnet 3.7, DeepSeek-chat V3, and Mistral Large—over a four-week period, discussing topics including climate change, math anxiety, and health-related misinformation. The dataset integrates demographic information, psychometric assessments, and multidimensional post-conversation subjective feedback. It offers the first longitudinal collection of persuasive dialogues enriched with contextual metadata and fine-grained subjective evaluations, enabling in-depth investigation into how AI systems influence human beliefs, anthropomorphic perceptions, and behavioral intentions over time, thereby establishing a foundational resource for research in persuasive human–AI interaction.
📝 Abstract
Talk2AI is a large-scale longitudinal dataset of 3,080 conversations (totaling 30,800 turns) between human participants and Large Language Models (LLMs), designed to support research on persuasion, opinion change, and human-AI interaction. The corpus was collected from 770 profiled Italian adults across four weekly sessions in Spring 2025, using a within-subject design in which each participant conversed with a single model (GPT-4o, Claude Sonnet 3.7, DeepSeek-chat V3, or Mistral Large) on three socially relevant topics: climate change, math anxiety, and health misinformation. Each conversation is linked to rich contextual data, including sociodemographic characteristics and psychometric profiles. After each session, participants reported on opinion change, conviction stability, perceived humanness of the AI, and behavioral intentions, enabling fine-grained longitudinal analysis of how AI-mediated dialogue shapes beliefs and attitudes over time.
Innovation
Methods, ideas, or system contributions that make the work stand out.
longitudinal dataset
human-AI interaction
persuasive dialogue
opinion change
large language models
🔎 Similar Papers
No similar papers found.
