AI-Augmented Work · Case Study
What "AI-assisted" actually means in course development
Allison Selby · createlearningexperiences.com
When I say a course is "AI-assisted," I don't mean the AI built it. I mean AI handled the scaffolding so I could focus entirely on the pedagogy. That distinction matters — and it's worth being specific about what it looks like in practice.
The problem it solves
The blank outline is one of the most time-consuming parts of course development — not because it's intellectually difficult, but because it requires translating SME notes, intake briefs, and business objectives into a logical learning sequence before anything else can happen. That translation work used to take two to four hours, and most of it wasn't instructional design. It was structural scaffolding.
AI closes that gap almost entirely. A draft course structure that used to take half a day now takes about 30 minutes to reach SME review. The difference isn't cutting corners — it's eliminating the blank page.
What AI does vs. what I do
AI handles
- Generating a draft module and lesson structure from intake notes
- Mapping content to Bloom's taxonomy levels
- Surfacing gaps between source material and stated objectives
- Producing first-draft narration scripts from approved outlines
- Suggesting assessment question formats by objective type
I handle
- Defining the learning objective and audience before AI is introduced
- Evaluating the AI draft against instructional design principles
- Sequencing for how adults actually learn — not just topic order
- SME relationship, accuracy review, and stakeholder sign-off
- All assessment design decisions and pass/fail criteria
The AI produces a structure. Whether that structure reflects sound instructional design — appropriate scaffolding, the right cognitive load, a learning arc that builds toward application — is entirely a human judgment call. That's the work.
The actual impact
Time to draft outline
30 min
Down from 2–4 hours. Draft ready for SME review same session.
Where time is spent
Pedagogy
Scaffolding is handled by AI. Human time goes to sequencing, Bloom's alignment, and gap analysis.
Outline quality
Equal or better
Gap-check step surfaces issues that previously slipped through to SME review.
SME review readiness
Same day
Outline reaches review stage in the same session as the needs analysis call.
The underlying principle
Instructional design has always been about spending time on decisions that require judgment — audience analysis, learning sequence, assessment design — not the administrative work of translating notes into a structure. AI doesn't change what good instructional design looks like. It removes the parts that were never really design work in the first place.
This workflow is one part of a broader AI-augmented production system developed across course development, assessment authoring, analytics, and coaching — documented in the workflow diagrams and prompt library elsewhere on this site. The common thread: AI handles the scaffolding. Humans handle the judgment.