Research on the Legal Liability Boundary of Algorithmic Recommendation: Taking Big Data Price Discrimination on E-Commerce Platforms as an Example

随着算法推荐技术在电商平台的广泛应用，以“大数据杀熟”为代表的算法滥用现象频发，暴露出法律规制滞后与技术权力扩张之间的结构性矛盾。本文聚焦电商平台差异化定价的认定困境，探讨算法推荐的法律责任边界问题。研究发现，现行法律框架固守“技术工具论”，难以有效应对算法的自主决策特性，导致消费者面临举证困境。法律法规分散、透明度缺失及举证责任失衡是规制失灵的核心症结。为此本文提出电商平台算法责任认定、构建初步证据–平台反证、整合法律法规体系三条重构路径，以平衡技术创新与权利保护，助力数字经济法治化。With the widespread application of algorithmic recommendation technology in e-commerce platforms, frequent occurrences of algorithm abuse such as “big data price discrimination” have exposed the structural contradiction between the lag in legal regulation and the expansion of technological power. This paper focuses on the identification dilemma of differential pricing on e-commerce platforms and explores the issue of the legal liability boundary of algorithmic recommendation. The research finds that the current legal framework adheres to the “technological tool theory” and has difficulty effectively addressing the autonomous decision-making characteristics of algorithms, leaving consumers facing difficulties in providing evidence. The core causes of the regulatory failure are the dispersion of laws and regulations, the lack of transparency, and the imbalance of the burden of proof. Therefore, this paper proposes three reconstruction paths: determining the algorithmic liability of e-commerce platforms, establishing a system of “Preliminary Evidence - Platform Counter-evidence”, and integrating the legal and regulatory system, so as to balance technological innovation and rights protection and promote the rule of law in the digital economy.