Can LLMs Infer Conversational Agent Users'Personality Traits from Chat History?

2026-03-31

AI Summary

This study investigates the privacy risks associated with inferring users’ personality traits from interactions with large language models (LLMs). Leveraging a real-world dataset of 62,090 messages from 668 users conversing with ChatGPT, the authors establish the first systematic framework for evaluating personality inference risk in authentic conversational settings. Using a RoBERTa-base model to perform ternary classification across the Big Five personality dimensions, they demonstrate that dialogues involving emotional relationships and self-reflection significantly heighten the risk of personality disclosure. Experimental results reveal that prediction accuracy for traits such as extraversion exceeds random baselines by up to 44%, providing compelling evidence that LLM-based conversational agents pose a tangible threat to personality privacy.

Abstract

Sensitive information, such as knowledge about an individual's personality, can be can be misused to influence behavior (e.g., via personalized messaging). To assess to what extent an individual's personality can be inferred from user interactions with LLM-based conversational agents (CAs), we analyze and quantify related privacy risks of using CAs. We collected actual ChatGPT logs from N=668 participants, containing 62,090 individual chats, and report statistics about the different types of shared data and use cases. We fine-tuned RoBERTa-base text classification models to infer personality traits from CA interactions. The findings show that these models achieve trait inference with accuracy (ternary classification) better than random in multiple cases. For example, for extraversion, accuracy improves by +44% relative to the baseline on interactions for relationships and personal reflection. This research highlights how interactions with CAs pose privacy risks and provides fine-grained insights into the level of risk associated with different types of interactions.

Problem

• personality inference
• conversational agents
• privacy risk
• chat history
• LLMs

Innovation

• personality inference
• conversational agents
• privacy risk
• LLM-based chat logs
• RoBERTa fine-tuning