第二次參加了「學與教博覽」，觀察到幾個值得分享的趨勢和變化。
HK EdTech AI Application and Teaching
(Text edited by ChatGPT 4o and translated into Traditional Chinese)
作為一名對教育應用充滿興趣的軟件架構師，我一直與教育工作者探討科技如何提升教與學的體驗。本週，我第二次參加了「學與教博覽」並進行展出，觀察到幾個值得分享的趨勢和變化。
從提示工程轉向 AI 工具
去年，討論的重點主要集中在提示工程——如何為 AI 工具編寫有效的提示。今年，焦點已轉向尋找能高效完成任務的工具，而不需要太多的輸入。現在的重點在於工具的可用性和成果，而不是與 AI 互動的技術細節。
大型語言模型的三層智能能力
AI 在教育中的應用可以概括為三個主要功能：
寫出更好的內容
快速分析非結構化文本
從數據中預測趨勢
大多數教師仍然專注於內容創作，利用 AI 來準備課程計劃、作業和教學材料。然而，分析非結構化數據和預測趨勢這兩個功能仍被認為過於技術性，與計算機科學相關。這些高階應用尚未被教育工作者廣泛理解或採用。
3.缺乏學生對 AI 工具的觀點
討論中一個明顯的缺口是學生的觀點。儘管教師和管理者正在探索 AI 的潛力，但對學生如何使用這些工具及其態度的理解有限。目前，學生主要將 AI 視為娛樂用途，而尚未完全將其用於自主學習。通過納入學生的聲音並探索他們使用 AI 的經驗，可以提供寶貴的見解，幫助這些工具更好地服務學習者。
如果您對這些想法感興趣，歡迎閱讀我關於「最近發展區」（ZPD）和元認知的博客，一起將科技融入教學中. https://www.hoplite-tech.com/blog
As a software architect with a deep interest in education, I have been collaborating with educators to explore how technology can enhance teaching and learning experiences. This week, while exhibiting at the Learning and Teaching Expo for the second year, I observed several trends and shifts worth sharing.
1. Shift from Prompt Engineering to AI Tools
Last year, much of the discussion centered around prompt engineering—how to write effective prompts for AI tools. This year, the focus has shifted toward finding tools that can efficiently complete tasks without needing extensive input. The emphasis is now on usability and outcomes rather than the technicalities of interacting with AI.
2. Large Language Model’s Three Levels of Intelligence Capabilities
AI in capability can be summarized into three main functions:
- Writing better content
- Analyzing unstructured text
- Predicting trends from data
Most teachers remain focused on the content-writing aspect, leveraging AI for lesson plans, assignments, and instructional materials. However, the other two functions—analyzing unstructured data and predicting trends—are still perceived as too technical and aligned with computer science. These advanced applications are not yet widely understood or embraced by educators. The next leap on edtech will depend on how teachers to use LLM beyond generating questions.
3. Lack of Student Perspectives on AI Tools
A noticeable gap in the discussions is the student perspective. While teachers and administrators explore AI's potential, there is limited understanding of how students are using these tools and their attitudes toward them. Currently, students mainly see AI in the context of entertainment rather than as a resource for self-directed learning. Bridging this gap by incorporating student voices and exploring their experiences with AI could provide valuable insights into how these tools can better serve learners.
Overall, the event highlighted the growing integration of AI in education but also underscored the need for broader adoption of advanced functions and a deeper exploration of student engagement with AI tools.
If you are interested in AI and edtech, I invite you to read my blogs on the Zone of Proximal Development (ZPD) and metacognition. Blog https://www.hoplite-tech.com/blog