技术范式转移下生成式AI搜索的垄断风险及其治理框架OACHSSCD

[Purpose]Generative AI search,powered by large language models,is driving a paradigm shift in information retrieval from"search engines"to"answer engines".Existing research has paid insufficient attention to the novel monopoly risks arising from this trans-formation.This paper systematically identifies the patterns of monopolistic conduct in generative AI search and their mechanisms of com-petitive harm,and proposes institutional responses tailored to the characteristics of this technological paradigm shift.[Method]A literature analysis is employed to trace the evolutionary trajectory of competitive paradigms in the search market.Case analysis and descriptive re-search methods are applied to examine the manifestations and regulatory dilemmas of monopolistic practices in generative AI search,inclu-ding self-preferencing,tying and bundling,compulsory content crawling,and exploitative pricing.[Result/Conclusion]The shift in competitive paradigms has reshaped the patterns of monopolistic conduct by search platforms,giving rise to dual risks of horizontal exclu-sionary abuse and vertical exploitative abuse.Antitrust governance of generative AI search is not a mere extension of traditional search en-gine competition issues,but rather a fundamental reexamination of competitive order in the search market under the context of technological paradigm shift.Antitrust law should move beyond the path dependence of ex post regulation and,informed by the Collingridge dilemma,establish a proactive and adaptive governance framework to achieve a dynamic equilibrium between fostering technological innovation and mitigating monopoly risks.