Balancing Artificial Intelligence and Human Expertise: Ideal Fact-Checking Strategies for Hard and Soft News

This study compares the effectiveness of artificial intelligence (AI)-aided versus human-assisted fact-checking for different types of news, namely hard and soft news. To this end, an online experiment was carried out separately in Korea and the United States, employing a 2 (news type: hard news vs. soft news) x 2 (fact-checker: AI vs. human experts) between-group factorial design. Findings suggest that AI's fact-checking, with its perceived objectivity and lack of bias, is potentially more effective for straightforward hard news, while human-assisted fact-checking appears to be more suitable for addressing misinformation within soft news, which involves more subjective aspects of information. Additionally, the study proposes that AI's verification of hard news and human experts' examination of soft news might trigger divergent indirect mechanisms-systematic processing driven by cognition or heuristic processing influenced by emotion-to counteract misinformation.