Empowering civic engagement in AI governance: A two-wave panel study on AI literacy and participatory governance of generative AI in China

The rapid development of generative artificial intelligence (AI) has posed significant regulatory challenges. To examine civic engagement in AI governance, this study conceptualizes participatory governance of generative AI and examines its communicative and cognitive antecedents with a focus on AI literacy. Drawing on the Orientation-Stimulus-Reasoning-Orientation-Response (O-S-R-O-R) framework, we conducted a two-wave panel survey of generative AI users in China. Using cross-sectional structural models and an autoregressive cross-lagged panel model (ACLPM), we investigated directional and reciprocal influences over time. Results reveal that news consumption about generative AI increases AI policy knowledge, while interpersonal discussion builds AI knowledge. These two types of knowledge mutually reinforce each other and play distinct roles. AI knowledge acts as an empowering orientation, increasing participatory governance by enhancing efficacy. Conversely, AI policy knowledge acts as a contextual orientation, directly motivating participation and future news consumption but does not directly increase efficacy. Instead, participation itself increases efficacy over time.The results highlight a critical distinction between “empowered” and “informed” participation in China’s state-led context. Furthermore, news consumption and participatory governance mutually reinforce one another. The findings demonstrate an upward, reinforcing spiral among media use, AI literacy, and participatory governance. This suggests that the O-S-R-O-R framework operates as a dynamic, reciprocal cycle rather than a static sequence. These findings further contribute to our understanding of the role of AI literacy in empowering and informing the public to engage with state-led AI governance.