电商场景下人工智能生成内容在商品图美化中的失真机理与治理研究
Mechanism and Governance of AIGC Product Image Beautification-Induced Distortion in E-Commerce
摘要: 随着生成式人工智能内容(Artificial Intelligence Generated Content, AIGC)相关技术应用于电商内容生产,商家可借助AIGC工具快速生成商品主图、AI模特图和场景展示图,降低了视觉内容生产成本并提升了营销效率。然而,AIGC商品图在颜色、材质、尺寸比例和场景效果上的过度美化,导致图片与实物不符、消费者预期落差及平台信任损耗等问题日益突出。本文以AIGC商品图美化失真为研究对象,从技术生成、商家选择、平台发布和消费者认知四个维度分析其形成机理。在此基础上,从增加实物比对、改变商家成本收益结构、将图片实物一致性嵌入审核与推荐、让售后反馈驱动前端纠偏四个方向提出治理路径,为电商平台在视觉营销效率与商品信息真实性之间建立可运行的平衡提供理论参考。
Abstract: With the application of Artificial Intelligence Generated Content (AIGC) technology in e-commerce content production, merchants can leverage AIGC tools to rapidly generate product main images, AI model images, and scene display images, thereby reducing visual content production costs and enhancing marketing efficiency. Yet the use of AIGC product images also raises concerns. When product images are excessively beautified in color, material texture, size proportion, scene presentation, or functional effect, the visual information conveyed to consumers may deviate from the actual product. Consequently, mismatches between displayed images and physical goods, inflated purchase expectations, and the weakening of platform trust have become more noticeable. Focusing on distortion caused by the beautification of AIGC product images, this paper analyzes its formation mechanism through four dimensions: technology generation, merchant selection, platform publication, and consumer cognition. On this basis, the study puts forward corresponding governance measures, including adding physical product comparison, adjusting merchants’ cost-benefit structure, embedding image-product consistency into review and recommendation systems, and allowing post-purchase feedback to support front-end correction. The purpose of this research is to offer theoretical reference for e-commerce platforms in coordinating visual marketing efficiency with the authenticity of product information.
文章引用:施季莹. 电商场景下人工智能生成内容在商品图美化中的失真机理与治理研究[J]. 电子商务评论, 2026, 15(11): 619-625. https://doi.org/10.12677/ecl.2026.15111280

参考文献

[1] Zhang, Y. and Liu, C. (2024) Unlocking the Potential of Artificial Intelligence in Fashion Design and E-Commerce Applications: The Case of Midjourney. Journal of Theoretical and Applied Electronic Commerce Research, 19, 654-670.
https://doi.org/10.3390/jtaer19010035
[2] 闫芳. 生成式多模态AI在电商智能营销中的融合应用与效能评估[J]. 营销界, 2025(10): 19-21.
[3] Hartmann, J., Exner, Y. and Domdey, S. (2025) The Power of Generative Marketing: Can Generative AI Create Superhuman Visual Marketing Content? International Journal of Research in Marketing, 42, 13-31.
https://doi.org/10.1016/j.ijresmar.2024.09.002
[4] 邵清华. 生成式人工智能对电子商务的促进作用研究[J]. 电子商务评论, 2025, 14(7): 1-6.
[5] Campbell, C., Plangger, K., Sands, S. and Kietzmann, J. (2022) Preparing for an Era of Deepfakes and AI-Generated Ads: A Framework for Understanding Responses to Manipulated Advertising. Journal of Advertising, 51, 22-38.
https://doi.org/10.1080/00913367.2021.1909515
[6] Seo, I.T., Liu, H., Li, H. and Lee, J. (2025) AI-Infused Video Marketing: Exploring the Influence of AI-Generated Tourism Videos on Tourist Decision-Making. Tourism Management, 110, Article 105182.
https://doi.org/10.1016/j.tourman.2025.105182
[7] 李宝麒. 银发经济视域下网络直播带货对县域老年群体消费行为的影响研究[J]. 老龄化研究, 2026, 13(5): 353-361.
https://doi.org/10.12677/ar.2026.135262
[8] Chaisatitkul, A., Luangngamkhum, K., Noulpum, K. and Kerdvibulvech, C. (2024) The Power of AI in Marketing: Enhancing Efficiency and Improving Customer Perception through AI-Generated Storyboards. International Journal of Information Technology, 16, 137-144.
https://doi.org/10.1007/s41870-023-01661-5
[9] 王张华, 李成成, 曾玉芝. 眼见未必为实——人工智能生成内容引发的社会信任风险及其化解策略[J]. 自然辩证法通讯, 2026, 48(7): 101-108.
[10] 浦天龙, 王既澎, 汤志伟, 等. 深度伪造信息传播的演化逻辑、潜藏风险与治理进路研究[J]. 情报科学, 2026, 44(7): 186-193.
[11] 张旭龙, 瞿晓阳, 谢骏飞, 等. 人工智能生成内容技术综述[J]. 大数据, 2026, 12(1): 146-173.
[12] Kramer, R.S.S., Jones, A.L., Fitousi, D. and Tree, J.J. (2025) AI-Generated Images of Familiar Faces Are Indistinguishable from Real Photographs. Cognitive Research: Principles and Implications, 10, Article 70.
https://doi.org/10.1186/s41235-025-00683-w