The spread of generative AI technology impacts copyright system in two key areas:first at the output end,the copyrightability and ownership of AI-generated content,second at the input end,the legal boundaries of large-scale use of others'works based on machine learning.On the first issue,there are both consensus and disputes among the representative countries in the world.While universal adherence and recognition that human involvement is a prerequisite for copyright,there are still cognitive disputes between users and designers over ownership.On the second issue,there is a big difference between AI technology-leading and catching-up countries in their institutional choice.Technology-leading countries rely more on case law tradition,waiting for the game between the traditional copyright industry and the emerging AI industry,to continue the approach of dealing with the Internet industry in the past,and to expect the industrial cooperation channels and development space based on full expression of their interests and positions.By contrast,technology catching-up countries are more inclined to legislative approach to support AI industry.Given China's current technological level and industrial status,it is necessary to integrate institutional experience of technology catching-up countries into the interpretation and adjustment of the local"three-step test",and breaking the limits of"nonprofit"and"moderateness"in machine learning.
万方数据
维普
