Use LLM-generated lessons within a human-AI collaborative workflow where educators review and refine AI drafts¶

Strategy · All strategies
Evidence · no claims cited

Description¶

A workflow in which AI-generated lesson content serves as a foundational draft that educators and instructional designers iteratively review and refine, with AI handling repetitive, time-intensive drafting tasks while human expertise ensures pedagogical alignment. The article states that LLM-generated lessons' "effectiveness is maximized within a human-AI collaborative workflow" and that educators "play a crucial role in reviewing and refining AI-generated content." Designers reported the AI drafts "can be used after a few rounds of modification."

Design Implications¶

Context¶

Requirements¶

• Educators or instructional designers review and refine the AI-generated draft content before publication
• Human verification of academic references, since generated references were often unreliable

Constraints¶

• The workflow was evaluated only through lesson designers' qualitative judgments, not through experiments measuring tutor learning outcomes

Target Learners¶

• novice human tutors teaching middle school mathematics online

Target Learning Goals¶

• effective tutoring practices
• high-quality scenario-based lesson content

Related Strategies¶

• Treat AI-generated lesson plans as drafts requiring teacher evaluation, adaptation, and pedagogical judgment
• Teachers iteratively refine LLM-simulated student ideas with classroom context before instructional use

Examples¶

-

Key Sources¶

• Jionghao Lin, Jiarui Rao, Sandy Yiyang Zhao, Yuting Wang, Ashish Gurung, Amanda Barany, Jaclyn Ocumpaugh, Ryan S. Baker, and Kenneth R. Koedinger. (2025). Automatic Large Language Models Creation of Interactive Learning Lessons. arXiv preprint. https://arxiv.org/abs/2506.17356