Scholarlink

<p>On social media, artificial intelligence (AI) increasingly curates content alongside social contacts. We examine whether social and algorithmic recommendations shape users’ perceived social norms around a moral issue and their intentions to engage with it. Drawing on theories of human–machine communication, human–AI interaction, and social norms, this experimental survey (<i>N</i> = 1,021) compares social, algorithmic, and popularity-based algorithmic recommendations (e.g., “most read”) in the context of digital immortality. Recommendation type did not affect perceived norms, and algorithmic appreciation did not moderate these effects. However, perceived social norms—especially norms attributed to one’s social environment—were positively associated with intentions to discuss and act on the issue. These findings suggest that recommendations do not deterministically exert normative influence; they, however, also point to the potential power of perceived norms in shaping engagement with emerging moral and technological issues. Future research should investigate the conditions under which algorithmic and social cues shape normative perceptions and help further clarify the role of AI-driven content curation in public discourse.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Recommended to you: an experimental study of normative influences from algorithmic and social recommendations on social media

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年3月12日

查看全文
  • Sarah Geber,
  • Lea Stahel

摘要

On social media, artificial intelligence (AI) increasingly curates content alongside social contacts. We examine whether social and algorithmic recommendations shape users’ perceived social norms around a moral issue and their intentions to engage with it. Drawing on theories of human–machine communication, human–AI interaction, and social norms, this experimental survey (N = 1,021) compares social, algorithmic, and popularity-based algorithmic recommendations (e.g., “most read”) in the context of digital immortality. Recommendation type did not affect perceived norms, and algorithmic appreciation did not moderate these effects. However, perceived social norms—especially norms attributed to one’s social environment—were positively associated with intentions to discuss and act on the issue. These findings suggest that recommendations do not deterministically exert normative influence; they, however, also point to the potential power of perceived norms in shaping engagement with emerging moral and technological issues. Future research should investigate the conditions under which algorithmic and social cues shape normative perceptions and help further clarify the role of AI-driven content curation in public discourse.