直播带货中AIGC虚拟主播的技术采纳与社会影响研究

Research on Technology Adoption and Social Impact of AIGC Virtual Streamers in Live Streaming E-Commerce

摘要: 在生成式人工智能深度嵌入数字经济的背景下，虚拟主播已成为直播电商产业降本增效的关键变量，但其实践中日益凸显出资本侧高效率预期与公众侧低社会接受之间的结构性矛盾。本研究立足于技术与社会互构的理论视角，通过构建“技术可供性–社会接受度”的整合性分析框架，深入解构了技术赋能下的商业实践如何引致信任异化。研究发现，AIGC虚拟主播在时间、形象与认知层面的技术可供性虽极大优化了商业效率，但其对工具的过度理性正侵蚀社会互动的根基。一方面，对“完美拟像”的统计学追求导致了真实性的本体论消解，具身互动的缺失在人机之间筑起了阻断情感流动的算法壁垒；另一方面，算法黑箱引发的责任分散效应，导致难以追责的“多重手”难题，进而从根本上瓦解了直播带货的制度性信任。基于此，本文提出从“效率优先”向“责任性设计”转型的治理路径，主张通过增强算法透明度、明确责任锚点及重塑人机协作模式，构建兼容技术效能与人文价值的负责任数字信任生态。

Abstract: In the context of the deep integration of generative artificial intelligence into the digital economy, virtual streamers have emerged as a critical variable for driving cost reduction and efficiency enhancement within the live streaming e-commerce industry. However, current practices increasingly highlight a structural contradiction between the high efficiency expectations of the capital sector and the low social acceptance of the public. Grounded in the theoretical perspective of sociotechnical construction, this study establishes an integrative analytical framework combining technology affordance and social acceptance to deconstruct how technology-empowered business practices induce trust alienation. The research reveals that while the technological affordances of AIGC virtual streamers in temporal, visual, and cognitive dimensions have significantly optimized commercial efficiency, the excessive instrumental rationality inherent in these tools is eroding the foundations of social interaction. On one hand, the statistical pursuit of a “perfect simulacrum” leads to an ontological dissolution of authenticity, where the absence of embodied interaction erects an algorithmic barrier that impedes emotional flow between humans and machines. On the other hand, the opacity of the “black box” algorithm triggers a diffusion of responsibility, resulting in the “problem of many hands”—a dilemma where accountability is obscured—thereby fundamentally dismantling the institutional trust essential to live commerce. Consequently, this paper proposes a governance transition from an “efficiency-first” paradigm to one of “responsible design”. It advocates for the construction of a responsible digital trust ecosystem that reconciles technical efficacy with humanistic values by enhancing algorithmic transparency, defining clear loci of responsibility, and reshaping human-machine collaboration models.