生成式人工智能垄断风险问题探析

Analysis on the Monopoly Risk of Generative Artificial Intelligence

随着生成式人工智能的广泛应用,将引发模型共谋、数据剥削、歧视性许可和生态封锁等多重垄断风险,同时在“资本-技术”和“算法-数据”双循环中形成结构壁垒和生态垄断,成为人工智能时代的新型数字基础设施。现行反垄断法无法有效应对生成式人工智能带来的新型垄断风险。针对这种新型的垄断样态,有必要在人工智能“三要素”逻辑基础上,传统反垄断规制框架下,细化规则,完善生成式人工智能新型垄断规制路径。

With the widespread application of generative artificial intelligence,it will trigger multiple monopoly risks such as model collusion,data exploitation,discriminatory licensing and exclusive licensing concentration,and at the same time form structural barriers and ecological monopolies in the dual circulation of“capital-technology”and“algorithm-data”,becoming a new digital infrastructure in the era of artificial intelligence.The current antitrust law cannot effectively deal with the new monopoly risks brought by generative artificial intelligence.In response to this new type of monopoly,it is necessary to refine the rules and improve the new monopoly regulation path of generative artificial intelligence based on the logic of the“three elements”of artificial intelligence and under the traditional antitrust regulatory framework.