Nudging Perceived Credibility: The Impact of AIGC Labeling on User Distinction of AI-Generated Content

The rapid advancement of generative artificial intelligence (AI) has made AI-generated content (AIGC) increasingly prevalent. However, misinformation created by AI has also gained significant traction in online consumption, while individuals often lack the skills and attribues needed to distinguish AIGC from traditional content. In response, current media practices have introduced AIGC labels as a potential intervention. This study investigates whether AIGC labels influence users’ perceptions of credibility, accounting for differences in prior experience and content categories. An online experiment was conducted to simulate a realistic media environment, involving 236 valid participants. The findings reveal that the main effect of AIGC labels on perceived credibility is not significant. However, both prior experience and content category show significant main effects ( P < .001), with participants who have greater prior experience perceiving nonprofit content as more credible. Two significant interaction effects were also identified: between content category and prior experience, and between AIGC labels and prior experience ( P < .001). Specifically, participants with limited prior experience exhibited notable differences in trust depending on the content category ( P < .001), while those with extensive prior experience showed no significant differences in trust across content categories ( P = .06). This study offers several key insights. First, AIGC labels serve as a viable and replicable intervention that does not significantly alter perceptions of credibility for AIGC. Second, by reshaping the choice architecture, AIGC labels can help address digital inequalities. Third, AIGC labeling extends alignment theory from implicit value alignment to explicit human–machine interaction alignment. Fourth, the long-term effects of AIGC labels, such as the potential for implicit truth effects with prolonged use, warrant further attention. Lastly, this study provides practical implications for media platforms, users, and policymakers.