The training and application of generative artificial intelligence large language models have caused large-scale copyright collective lawsuits in the United States and other regions,and copyright issues have become the primary obstacle to the development of large language models.In terms of risk types,the copyright class action lawsuit of the large language model has triggered accompanying legal risks,economic risks,and innovation risks.The three important measures taken by the United States to respond to collective lawsuits over copyright in large language models are to consolidate social consensus through open hearings,buffer copyright conflicts through active judiciary,and eliminate copyright risks through industry autonomy,reflecting the innovative governance approach of prioritizing innovation.The big language model in our country is currently in a critical stage of research and application,and is facing similar copyright risks.The experience of the United States serves as a valuable reference for copyright governance in China's large language models.Our country's copyright governance measures should adhere to the value orientation of"innovation first"at the macro level,adopt a governance model of"diverse collaboration"at the meso level,and introduce the type of"reasonable use of artificial intelligence training"at the micro level.
关键词
生成式人工智能/大语言模型/版权集体诉讼/美国经验
Key words
Generative Artificial Intelligence/Big Language Model/Copyright Collective Litigation/Innovation Priority
万方数据