Job Aid · AI Content Evaluation · Editorial Judgment
Evaluating
AI-Generated Content
A one-page framework for judging AI-drafted content against explicit criteria, writing a rationale that holds up, and catching what a fluent-sounding draft can still get wrong.
The Job Aid
Turning judgment into a repeatable system
Using AI to draft content is routine now. The differentiator is whether the person directing it can tell good output from confident-sounding output, and say precisely why, before it ships rather than after.
This job aid formalizes how I review AI-drafted content: four criteria in priority order, a checklist for catching what a fluent read misses, and a method for writing a rationale that leads with the decisive factor instead of building up to it.
What It Covers
Five sections, one page
Why it matters
The case for critical review over accepting a fluent first draft.
Four criteria, ranked
Accuracy and instruction adherence come first and override clarity and consistency when content fails either.
A repeatable checklist
Terminology, transitions, unstated assumptions, references, and constraint compliance.
Rationale method
Lead with the decisive factor, one piece of evidence, then stop. Write to the limit, not under it.
Worked example
Two draft responses compared against the criteria, with a properly constrained rationale.
Built from real practice
Formalized from reviewing AI-drafted instructional content at LearningMate and in portfolio work.
Interested in working together?
I'm currently available for remote instructional design and eLearning development work.