This study examines university students levels of satisfaction with generative artificial intelligence (AI) tools and traditional search engines as academic information sources. An electronic survey was distributed to students at U.S. universities in late fall 2025, with 236 valid responses received. In addition to demographic information about respondents, frequency of use and levels of satisfaction with both generative AI and traditional search engines were measured. Principal components analysis identified distinct constructs of satisfaction for each information source, while k-means cluster analysis revealed two primary student groups: those highly satisfied with search engines but dissatisfied with AI, and those moderately to highly satisfied with both. Regression analysis showed that frequency of use strongly predicts satisfaction, with international and undergraduate students reporting significantly higher satisfaction with AI tools than domestic and graduate students. Students generally expressed higher levels of satisfaction with traditional search engines over generative AI tools. Those who did prefer AI tools appear to see them more as a complementary source of information rather than a replacement for other sources. These findings stress evolving patterns of student information seeking and use behavior and offer meaningful insights for evaluating and integrating both traditional and AI-driven information sources within higher education.

翻译：本研究探讨了大学生对生成式人工智能工具与传统搜索引擎作为学术信息来源的满意度水平。一项电子调查于2025年秋末在美国多所大学的学生中展开，共回收236份有效问卷。除受访者的人口统计信息外，研究还测量了生成式人工智能与传统搜索引擎的使用频率及满意度水平。主成分分析识别出针对每种信息来源的独立满意度构念，而k-means聚类分析则揭示出两个主要学生群体：对搜索引擎高度满意但对AI不满意的群体，以及对两者均持中度至高度满意态度的群体。回归分析表明，使用频率能强烈预测满意度，其中国际学生与本科生对AI工具的满意度显著高于本国学生与研究生。总体而言，学生对传统搜索引擎的满意度普遍高于生成式人工智能工具。那些更偏好AI工具的学生似乎更多将其视为补充性信息来源，而非其他来源的替代品。这些发现凸显了学生信息检索与使用行为的演变模式，并为高等教育领域评估与整合传统及AI驱动的信息来源提供了重要启示。