With the wide application of generative artificial intelligence (AIGC) technology in the fields of text generation, image synthesis and voice forgery, the efficiency of content production has been significantly improved, but it also brings new social governance challenges such as the proliferation of false information and the intensification of public opinion manipulation. This paper focuses on the problem of false content generation in the context of AIGC, and analyses the types of risks, dissemination paths and public perception effects triggered by the problem in the public opinion arena, taking into account typical cases. It further explores the roles and limitations of the government and platforms in risk identification, information verification, algorithmic traceability and response mechanism, and constructs a triadic interaction model of "generation risk-public perception-governance mechanism". The study proposes to establish a multifaceted and coordinated algorithmic governance system, including the construction of a technical early warning mechanism, the implementation of AI content traceability and labelling system, the enhancement of public digital literacy, and the compaction of platform responsibility. The study shows that the risk caused by AIGC is highly realistic and rapidly spreading, and the traditional means of public opinion governance urgently needs to be transformed to an algorithm-centred systematic synergistic mechanism. This paper is of great theoretical significance and practical value for reconstructing the public trust mechanism and promoting the modernisation of national public opinion governance.
Qin et al. (2025) studied this question.