Using an AI Evidence Logbook in Course Projects
An AI evidence logbook helps students document prompts, outputs, revisions, evaluation notes, and responsible-use decisions throughout a project.
When students use AI tools in course projects, the main question should not be only whether AI was used. The better question is whether the student can document how AI shaped the work and how the output was evaluated.
An AI evidence logbook gives students a lightweight structure for that documentation.
Recommended logbook fields
For each meaningful AI interaction, students record:
- date and task goal
- prompt or prompt summary
- model/tool used
- useful output excerpt
- errors or limitations found
- revision made by the student
- verification source
- decision to accept, modify, or reject
This makes AI use visible without turning the class into surveillance.
Assessment use
The logbook helps lecturers assess process quality. A student who records weak outputs and explains corrections may show stronger learning than a student who submits polished work with no evidence trail.
Learning outcome
Students learn that responsible AI use requires judgment. The evidence logbook turns that judgment into something reviewable, teachable, and improvable.