Large Language Models Can Assist in Generating Personality-Matched Anti-Fraud Voice Warnings in the Chinese Context

With the rise of AI-generated content, personalized interventions have become increasingly feasible in high-risk decision-making contexts such as online fraud. This research examined whether large language models (LLMs), exemplified by DeepSeek, can generate personality-matched voice warnings to enhance anti-fraud persuasion in the Chinese context. Across two studies, warnings were tailored to Big Five personality traits by manipulating either text content only (Study 1, N = 296) or both text and vocal parameters (Study 2, N = 191). Study 1 showed that personality-matched warnings significantly increased perceived warning effectiveness, with the clearest effect for Openness. However, Study 2 found no significant personality-matching effects when both text and vocal parameters were generated by the LLM. These findings suggest that LLMs may support personality-based warning generation under controlled text-only conditions, but their effectiveness remains limited when extended to multidimensional voice-warning design.