生成式人工智慧系統與網路搜尋之耳鳴資訊品質：專家評比比較研究

// Quality of Tinnitus Information From Generative AI Systems and Web Search: An Expert‑Rated Comparative Study

本研究比較六種生成式人工智慧系統與Google搜尋的耳鳴相關資訊品質｜OpenEvidence在準確性、清晰度、相關性、完整性與實用性上表現最佳，優於其他AI及搜尋結果，但整體完整性偏低｜研究發現高品質資訊可讀性較差，顯示AI提供較佳資訊需結合臨床人員監督，確保品質與患者理解間取得平衡。

// This study compared tinnitus-related information quality from six generative AI systems and Google Search using expert ratings. OpenEvidence ranked highest in accuracy, clarity, relevance, completeness, and usefulness, outperforming other AI and search results, though completeness was generally low. Readability was inversely related to quality, with higher-quality responses being less accessible. The results suggest AI often provides better information than web search but highlight the need for ongoing evaluation and clinician oversight to balance quality and patient comprehension.

Published 2026年7月20日