| Home

Overview


Original Research

THE STRUCTURAL IMBALANCE OF COMPETITIVE DYNAMICS IN E-COMMERCE PLATFORMS UNDER AI AGENT INTERVENTION AND THE RECONSTRUCTION OF LEGAL REGULATION

TING HOU and XIANG LI.

Vol 21, No 06 ( 2026 )   |  DOI: 10.5281/zenodo.20588248   |   Author Affiliation: 2nd Year Undergraduate Student, The School of Publishing, University of Shanghai For Science and Technology, Majoring in Publishing And Editing 1, Lecturer and Master’s Supervisor, Department of Publishing and Digital Communication, School of Publishing, University of Shanghai for Science and Technology 2.   |   Licensing: CC 4.0   |   Pg no: 18-29   |   Published on: 08-06-2026

Abstract

Generative artificial intelligence–driven AI agents, characterized by cross-platform orchestration, autonomous decision-making, and task-planning capabilities, are fundamentally reshaping the competitive structure of e-commerce platforms. The traditional decision-making pathway—user, platform, and product—is increasingly transitioning toward a distributed architecture involving users, AI agents, and multiple platforms. In this emerging configuration, platform-centric advantages in entry control, traffic allocation, and data aggregation are significantly weakened, while the core of market competition shifts from intra-platform dynamics to algorithm-mediated intermediation. This paper conceptualizes AI agents as quasi-market actors and examines their structural challenge to platform governance power. It identifies several systemic tensions, including conflicts over gateway control, frictions between data governance and personal information protection, distortions in algorithmic decision authority, and the fragmentation of multi-actor responsibility frameworks. These dynamics collectively contribute to a reconfiguration of digital market power and regulatory uncertainty. To address these challenges, the paper proposes a set of governance reconstruction pathways. These include the establishment of auditable API access systems for platform entry control, the development of behavioral data authorization pools to regulate data circulation, the implementation of recommendation influence labeling mechanisms to enhance algorithmic transparency, and the construction of a segmented liability framework along the decision chain. This framework differentiates responsibilities across data generation, recommendation processing, and transaction execution stages. Through these institutional designs, the study aims to preserve technological innovation while redefining the power boundaries among platforms, AI agents, and users, thereby safeguarding fair competition in digital markets.


Keywords

AI Agents; e-Commerce Platforms; Algorithmic Governance; Data Circulation; Legal Liability Allocation.