Author(s)

Yantong Lin, Yulin Zhao

Corresponding Author

Yulin Zhao

Abstract

With the rise of generative artificial intelligence, multimodal news production has entered a new stage of automated and intelligent expression. However, emotional distortion problems caused by excessive algorithmic generation and style transfer threaten news credibility and public trust. This study examines the mechanism of emotional expression distortion in AI-driven multimodal news and explores credibility restoration strategies. By analyzing the causes of distortion from content, technology, and communication perspectives, and constructing governance and repair pathways, this research aims to provide theoretical support and practical reference for improving the quality and trustworthiness of news communication in the intelligent era.

Keywords

Generative AI; Multimodal News; Emotional Distortion; News Credibility; Trust Restoration