从再现到生成:AI影像、深度伪造与真实性政治(From Representation to Generation: AI Imagery, Deepfakes, and the Politics of Authenticity)

Read full text Summarise with VIRAT
Authors: Xiaofang Cai, Yuqing Liu
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The operational logic of artificial intelligence (AI) imagery constitutes a form of the “politics of authenticity.” Here, the “politics of authenticity” refers to the idea that the real is no longer a self-evident ontological given, but rather a symbolic system jointly produced and maintained by algorithms, power structures, and cultural frameworks. The visual logic of images has undergone a historical shift from representation to generation. In this transformation, images gradually detach from an indexical foundation and become “verisimilar” experiences predicted and generated through data and models. Once the real loses its self-evidence at the level of ontology, images increasingly function as governance resources that can be operationalized, produced, and distributed within structures of power. The ultimate aim of deepfakes is therefore not merely to deceive individuals, but to sustain a macro-structural order governed by algorithmic systems. From a global perspective, such algorithmic governance operates through pronounced inequalities. Western techno-capitalist power monopolizes generative sovereignty, allowing AI-generated imagery to perpetuate structural biases related to race, gender, and class in more concealed and automated forms. Consequently, it becomes necessary to cultivate a critical consciousness appropriate to the post-truth era, in which the “real” is understood as a dynamic political process that must be continually contested and reinterpreted.

Fields of Study & Subjects

Cultural Studies and Postmodernism Photography and Visual Culture Ethics and Social Impacts of AI

Publication details

Journal / source
Zenodo (CERN European Organization for Nuclear Research)
Publication date
1 February 2026
Language
EN