生成式人工智能领域的垄断风险及其治理路径  

Monopoly Risk and Its Governance Path in the Field of Generative Artificial Intelligence

作　　者：雷浩然 LEI Hao-ran(Law School of Civil,Business and Economics,China University of Political Science and Law,Beijing 100088,China)

机构地区：[1]中国政法大学民商经济法学院,北京市100088

出　　处：《中国流通经济》2025年第4期115-126,共12页China Business and Market

基　　金：北京市社会科学基金青年项目“反垄断法视野下的企业数据行为治理”(20FXC019)。

摘　　要：近年来,随着生成式人工智能不断催生新场景、新业态和新市场,世界各国纷纷出台支持人工智能发展的战略计划和政策措施,以期抢占未来发展的制高点。与此同时,生成式人工智能领域存在的垄断风险也引发各国的广泛关注。从实践来看,生成式人工智能领域的垄断风险主要表现为排他风险、剥削风险和共谋风险,即抑制竞争对手的市场进入或扩张、扭曲消费者对基础模型服务的选择以及行业内企业之间易于达成共谋。生成式人工智能领域垄断风险的形成原因与相关市场高度集中、商业模式复杂多样和反垄断监管相对滞后等因素密切相关。治理生成式人工智能领域垄断风险的主要路径有三:一是合规倡导路径。引导生成式人工智能企业进行反垄断合规,鼓励生成式人工智能企业结合自身经营状况进行垄断风险的分析评估和分级管理,并加强与反垄断执法机构的沟通和协调。二是行业监管路径。结合生成式人工智能领域的产业发展现状和国际竞争情况制定行业监管政策,实现不同行业监管部门之间的监管联动。三是反垄断执法路径。开展生成式人工智能领域的反垄断执法,提高反垄断法律规范在生成式人工智能领域的适用性,实现促进创新和保护竞争的有机协调。其中,合规倡导侧重于生成式人工智能企业的内部自治,行业监管和反垄断执法侧重于生成式人工智能企业的外部管控。这三条路径相互配合、密切协作,能够形成应对生成式人工智能领域垄断风险的综合性治理体系,从而有效保护生成式人工智能领域的市场公平竞争。In recent years,with the continuous emergence of new scenarios,formats,and markets through generative artificial intelligence,countries around the world have introduced strategic plans and policy measures to support the development of artificial intelligence,in order to seize the high ground of future development.At the same time,the monopoly risk in the field of generative artificial intelligence has also attracted widespread attention from various countries.From a practical perspective,the monopoly risk in the field of generative artificial intelligence mainly manifests as exclusivity risk,exploitation risk,and collusion risk,namely,inhibiting competitors'market entry or expansion,distorting consumers'choices of basic model services,and facilitating collusion among enterprises in the industry.The formation of monopoly risks in the field of generative artificial intelligence is closely related to factors such as highly concentrated related markets,complex and diverse business models,and relatively lagging competition governance.There are three main paths to address monopoly risks in the field of generative artificial intelligence:first,the compliance advocacy path,guiding generative artificial intelligence enterprises to comply with anti-monopoly regulations,encouraging them to analyze,evaluate,and classify monopoly risks based on their own business conditions,and strengthening communication and coordination with anti-monopoly law enforcement agencies;second,the industry regulatory path,which combines the current development status and international competition in the field of generative artificial intelligence to formulate industry regulatory policies and achieve regulatory linkage between different industry regulatory departments;and third,the path of anti-monopoly law enforcement,carrying out anti-monopoly law enforcement in the field of generative artificial intelligence,improving the applicability of anti-monopoly legal norms in the field of generative artificial intelligence,and achieving organic coordinat

关 键 词：生成式人工智能 基础模型 ChatGPT DeepSeek 垄断风险 

分 类 号：F203[经济管理—国民经济]