Generative-AI-assisted content creation with domain-expert validation for adaptive tutoring systems¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The G4L authors used a generative AI assistant to build the knowledge base rapidly from existing documentation: "the Generative AI (GenAI) assistant automatically generated the prerequisite relational structure of knowledge units, alongside the initial submaterials for each unit." Quiz questions and answers were likewise generated from the established content structure. All machine-generated materials were then "subjected to a refinement process by a domain expert to achieve its final, pedagogically sound form," realizing a teacher-in-the-loop approach in which educators curate and validate AI-generated outputs.

Design Implications¶

Context¶

Requirements¶

• Existing documentation of the educational material as input, and a domain expert to review and validate all generated content

Constraints¶

• The process is semiautomated: AI output is a draft that requires expert refinement before pedagogical deployment

Target Learners¶

• instructors and domain experts authoring adaptive tutoring content in dynamic domains

Target Learning Goals¶

• rapid establishment of a coherent and comprehensive knowledge base for adaptive learning environments

Related Strategies¶

• Computing instructors should approach GenAI slide generation as programmers: work iteratively with coding assistants and text-based slide toolchains
• Benchmark AI assistants on specialized educational tasks before classroom deployment
• Human-in-the-loop validation workflow: hold all AI-generated teaching content in unreviewed status until a teacher manually approves it

Examples¶

-

Key Sources¶

• Csépányi-Fürjes, L., & Kovács, L. (2026). Intelligent tutoring in dynamic domains: a graph-based system for comparative analysis of adaptive algorithms with intuitionistic fuzzy logic and forgetting. Education Tech Research Dev. https://doi.org/10.1007/s11423-026-10639-6