关键词: 生成式人工智能, 治理实践, 欧美比较, 人工智能治理, 人工智能伦理

Abstract:

[Purpose/Significance] Generative artificial intelligence (GAI) is currently advancing at an astonishing pace. GAI has unleashed remarkable potential in various fields and is significantly fueling social and economic development. However, this rapid progress has also given rise to a plethora of complex issues, including but not limited to data security breaches, privacy violations, the spread of false information, and intellectual property infringements. Existing research primarily focuses on the governance of AI in general, leaving a gap in in-depth exploration of GAI. This study aims to fill this void by meticulously comparing the governance approaches of Europe and the United States in the realm of GAI. Through this comparison, the study aims to provide valuable insights for China to refine its own governance system. This is not only crucial for China's domestic technological development and social stability but also plays a pivotal role in promoting the harmonization of the global governance framework for GAI. [Method/Process] This research adopts a multi-faceted approach. It commences with a comprehensive review of relevant literature, gathering insights from a wide range of academic sources to understand the current state-of-the-art in GAI governance in Europe and the United States. Additionally, it deploys the case-study method, examining real-world examples such as the development of OpenAI's GPT series in the US and the implementation of the EU's AI Act. By analyzing these cases, it can vividly illustrate the practical implications and impacts of different governance strategies, thus enabling a more in-depth and accurate comparison. [Results/Conclusions] We found that the European Union adopts a regulatory path centered on data protection and ensures the fairness and sustainability of technological development through a strict legal framework. However, this strong regulatory model may stifle innovation vitality to some extent. The United States adopts a governance model oriented towards market accountability, emphasizing technological innovation leadership and free development. It stimulates market vitality through industry self-discipline and flexible regulation, but there is a hidden danger of insufficient ethical risk control. Based on these findings, this paper recommends that China adopt a balanced approach. China should integrate elements of both the U.S. and E.U. models to foster innovation while ensuring ethical and legal compliance. Future research could explore ways to adapt these governance models to emerging trends such as integrating GAI with other emerging technologies and addressing the unique governance challenges posed by cross-border data flows.

Key words: generative AI, governance practice, comparison between Europe and the United State, AI governance, AI ethics