人工智慧時代下的科學探究與實作之評量與其展望：評量什麼（what）？如何評量（how）？

Assessing Scientific Inquiry and Practice in the Era of Artificial Intelligence: What Should We Assess, and How?

人工智慧（Artificial Intelligence, AI）的迅速發展，為科學教育中的探究與實作評量帶來契機與挑戰。本文聚焦機器學習（Machine Learning, ML）與生成式人工智慧（Generative Artificial Intelligence, GenAI）在科學探究評量中的應用，探討如何結合這些技術以推動評量創新，並特別著重於中小學階段的教育實踐。文章基於探究為本的學習架構，分析AI於「定題」、「概念化」、「調查」、「討論」與「結論」等探究面項的潛在應用。然而，AI評量也面臨資料偏見、缺乏多元性支持等挑戰。據此，本文提出一個整合ML與GenAI的動態評量系統以驅動教學與課程，並揭示教師與研究者在AI時代需具備的素養發展策略。

The rapid advancement of artificial intelligence (AI) presents both opportunities and challenges for assessing scientific inquiry and practice. This study examined the integration of machine learning and generative AI in inquiry-based assessments in K-12 education. Grounded in an inquiry-based learning framework, the study examined AI’s role across key inquiry components: problem formulation (orientation), conceptualization, investigation, discussion, and conclusion. AI-driven assessments face challenges, including data biases and limitations in capturing the diversity and complexity of scientific inquiry. To address these challenges, this study proposed a dynamic assessment system that leverages machine learning and generative AI to enhance teaching and curriculum development. Additionally, the study highlighted the essential competencies that educators and researchers must develop to effectively integrate AI into education.