算法推荐模式下网络平台的行为责任研究——以A公司为例
Research on Behavioral Responsibility of Network Platform under Algorithm Recommendation Model—Taking Company A as an Example

在短视频的间接侵权之中，作为短视频网络平台的核心技术方法——个性化推荐算法的介入，带来了更高的侵权风险和侵权形式。对于平台援引“避风港原则”的正当性也备受质疑。其主要原因是算法推荐模式下，网络平台的主观过错认定标准模糊。本文重点分析网络平台的主观过错责任，其通过理论与司法实践为依据去合理拓宽“明知”的适用范围，以此提高平台间接侵权的认定的可能性，更有利于平衡平台与权利人之间的关系。其次对平台的惩罚性赔偿适用问题，为了避免加重平台的经济负担，应该严格按照“故意 + 情节严重”构成要件。而且从利益平衡和技术可能性角度来看，数字推荐模式下的平台应该承担更高的注意义务，采取必要措施去预防和制止侵权行为的发生。
In the indirect infringement of short video, the intervention of personalized recommendation algorithm, which is the core technology of short video network platforms, brings higher infringement risk and infringement forms. The legitimacy of the platform’s invocation of the “safe haven principle” has also been questioned. The main reason is that the subjective fault identification standard of network platform is fuzzy under the algorithm recommendation model. This paper focuses on the analysis of the subjective fault liability of the network platform, which reasonably broadens the scope of application of “knowing” through theory and judicial practice, so as to improve the possibility of determining the indirect infringement of the platform, and is more conducive to balancing the relationship between the platform and the right holder. Secondly, the application of punitive damages to the platform, in order to avoid increasing the economic burden of the platform, should be strictly in accordance with the “intentional + serious circumstances” that constitute the elements. Moreover, from the perspective of interest balance and technical possibility, the platform under the digital recommendation model should bear a higher duty of care and take necessary measures to prevent and stop the occurrence of infringement.