Before You Start Orientation

A Learner's Permit authorizes practice. The 15 exercises in this scaffold are structured practice. Completing them — with full documentation — makes a student eligible to sit for the General AI License exam.

Exercise 15 is the exam. There is no separate examination event. A General AI License is awarded when Exercise 15 is accepted by an instructor.

[TOPIC] is a placeholder throughout. The instructor replaces it with a specific subject. The scaffold works for any discipline.

Prerequisites — Toolkit Setup

Three things must be functional before Exercise 1.

1. A model

Access to at least one large language model — ChatGPT, Claude, Gemini, or equivalent. More than one model will be used before Exercise 15.

2. A transcript exporter

A tool that saves the full text of a conversation to a file. The instructor specifies which exporter to use and how to configure it.

Filename convention: all lowercase, hyphens, no spaces. Set the default filename to include name and model: smith-claude-. Rename each conversation to reflect the exercise before exporting: lp-ex1-smith.txt, lp-ex2-smith.txt, and so on.

3. A workbench

A cloud folder with managed access and shareable links — SharePoint, OneDrive, Google Drive, or institutional equivalent. Create a root folder named ai-literacy. Inside it, create five subfolders:

ai-literacy/
  ├── transcripts/   ← exported conversations
  ├── artifacts/    ← AI-assisted documents and outputs
  ├── synthesis/    ← synthesis documents
  ├── author/      ← human-only writing (see below)
  └── zip/         ← dated workbench snapshots

/author discipline: Documents in /author are written without opening a model. Plain text editor only. Pre-closing reflections, peer critiques, task notes written before any AI consultation — anything where the thinking is unmediated — goes here. If a model was open when the document was written, it does not belong in /author.

The workbench is portable and model-agnostic. It grows across all 15 exercises and beyond, into whatever coursework follows. Build it once. Maintain it continuously.

Three Practices RTW Framework

RTW is the spine of every exercise:

  • Read — read what the model generates with attention. Consider what is asked of the model to read, how the reading task is framed, and what it is asked to read for. Reading is bidirectional.
  • Think — evaluate what the model generates. Agree, disagree, push back, extend. Thinking happens in the conversation — it shapes the next prompt.
  • Write — writing in this course means prompting. A prompt is a piece of writing that causes the model to generate text toward a specific outcome. The model's response is also writing. Prompting is how generation is directed.

The artifact is the model's final response. The student shapes through prompting. The model writes the final response. The artifact is that response, copied out of the conversation and saved to /artifacts.

Every exercise from Exercise 2 forward produces an artifact and a transcript. The artifact goes to /artifacts. The full exported conversation goes to /transcripts. Both are the work.

Phase I — Learning to Drive Exercises 1–5
Phase I · Exercises 1–5

Learning to Drive

Learn the instrument before playing it.
01 First Conversation

Have a real conversation with the model about [TOPIC]. Minimum 3 prompts. Follow what interests you. Push back on something. Ask a follow-up.

Read what the model produces. Note where it is confident. Note where skepticism is warranted.

No artifact in Exercise 1.

Transcript: rename conversation lp-ex1-[name] before exporting. Save to /transcripts.
02 Compression and Progression
Prompt
Summarize [TOPIC] in 25 words.

That is the baseline. Now expand — not a new summary, a deeper version of the same one:

Prompt
Expand that to [multiplier × 25] words. Not a new summary — a deeper version. Someone who read the 25-word version should feel like they're going further, not starting over.

Choose a multiplier of at least 3. Repeat two more times with the same multiplier. Example: 25 → 75 → 225 → 675. Or 25 → 100 → 400 → 1600. The pattern matters more than the specific numbers.

Goal: by the longest version, the ideas should feel familiar — recognized rather than newly encountered.

Prompt
At which level did [TOPIC] start to feel familiar to me, based on what I've said? Is this a useful way to learn this subject, or does it flatten things I'd rather encounter at full depth?
Prompt
Generate something useful specifically for me — a personal reference, a key question to hold onto, a way of remembering this material, or something else that fits how I've described my learning. Make it for me, not a general reader.
Artifact: the final response. Save to /artifacts as lp-ex2-[name].txt.
Transcript: lp-ex2-[name].txt/transcripts.
03 Push Back

Upload the Exercise 1 transcript to a new conversation.

Prompt
This is a transcript of a conversation I had with an AI about [TOPIC]. Please read it and repeat the final response in that transcript.

Read that response again — the model's unguided output from before anything was shaped. Find one claim to push on.

Choose any two of the following moves:

  • "I don't understand this." Ask for a different explanation. Don't accept the first re-explanation. Push again.
  • "My experience tells me otherwise." State what observation or prior learning contradicts the claim.
  • "This reading says something different." Paste a contradicting sentence from any source.
  • "Here's another model's response." Take the same prompt to a different AI, copy that response, ask this model to respond to it.

Note whether the model defends its position or concedes because of pressure. Agreeable is not the same as accurate.

Prompt
Summarize this exchange in 3–4 sentences: what I challenged, how you responded, and what the disagreement or clarification revealed about [TOPIC].
Artifact: that summary. Save as lp-ex3-[name].txt/artifacts.
Reflect
Before closing the conversation, write a few sentences in the chat — informal, stream of consciousness. Did this work? What surprised you? What were you thinking? Don't edit it. These reflections are part of the transcript.
Transcript: lp-ex3-[name].txt/transcripts.
04 Trust Audit
Prompt
What are the limits of your knowledge about [TOPIC]? Where should I not trust you?

Read carefully. Models are sometimes honestly uncertain — and sometimes evasive in ways that sound like honesty.

Does the model's list match the skepticism that emerged in Exercises 1–3? Are there gaps it didn't flag?

Write 3–5 sentences in the chat — a personal assessment of what to trust and what not to trust this model to do with [TOPIC].

Prompt
Incorporate my response into a short artifact titled 'Trust Audit: [TOPIC].' Capture my assessment and add anything you think I should also be skeptical about that I haven't mentioned.
Artifact: Trust Audit. Save as lp-ex4-[name].txt/artifacts.
Reflect
Write informally in the chat. What did the model say that was surprising? What did you expect it to admit that it didn't? What do you actually trust it to do?
Transcript: lp-ex4-[name].txt/transcripts.
05 Deep Conversation — 15 Minutes

Set a timer for 15 minutes. Have a sustained conversation with the model about [TOPIC]. Follow threads. Disagree. Ask it to explain things differently. Go somewhere unplanned.

Don't start over if it goes sideways. The mess is part of the record.

When the timer ends, stop.

Reflect
Write directly in the chat before the closing prompt — informal, unedited: What surprised you? Where was the model most useful? Least? What question didn't get asked? What would you do differently? This reflection is now part of the conversation. The closing prompt can draw on it.

Then choose one closing prompt:

Option A
Write a 150-word summary of what I engaged with in this conversation, organized as 3–4 bullet points — like a slide I could review later.
Option B
Give me the 3 most important questions this conversation raised that I haven't answered yet.
Option C
Write a 100-word paragraph capturing the central insight of this conversation and what I should explore next.
Option D
Give me a metaphor or analogy that captures the most important idea we discussed about [TOPIC].
Artifact: the response to the chosen option. Save as lp-ex5-[name].txt/artifacts.
Transcript: lp-ex5-[name].txt/transcripts.

From Exercise 5 forward, personal reflection comes before the closing prompt. Thinking shapes the artifact.

Phase II — Building Skill Exercises 6–10
Phase II · Exercises 6–10

Building Skill

Drive it where you want to go.
06 Fork — New Model, Full Context

Open a new conversation in a different model than Exercise 5. Upload the Exercise 5 transcript.

Prompt
I'm uploading a conversation I had with another AI about [TOPIC]. Please read it carefully. I'll ask you to continue from where it left off.

Once it confirms, ask the question that didn't get addressed in Exercise 5. Ask at least 3 follow-up questions.

Reflect
Write in the chat before the closing prompt. What felt different about this model — tone, assumptions, depth, behavior? What did it do with the context it was given?
Prompt
Summarize what this conversation added to my understanding of [TOPIC] that the previous conversation did not. Be specific about what was new, different, or contradictory.
Artifact: that summary. lp-ex6-[name].txt/artifacts.
Transcript: lp-ex6-[name].txt/transcripts.
07 Same Content, Three Audiences
Prompt
Explain [TOPIC] three times: once for a curious 10-year-old, once for a first-year college student, once for a working professional in the field.

Read all three. Note where facts change versus where only framing changes. Note what the professional version assumes the reader already knows.

Use one of the two moves from Exercise 3 to push on one of the versions.

Reflect
Which version would you actually read? Which version is most honest about complexity?
Prompt
Write a short artifact analyzing the structural differences between these three versions — tone, assumptions, density. Not the content: the architecture. Note which version served me best as a learner and why.
Artifact: that analysis. lp-ex7-[name].txt/artifacts.
Transcript: lp-ex7-[name].txt/transcripts.
08 First Reading

The instructor provides a reading related to [TOPIC]. Upload the PDF.

Prompt
I'm uploading a reading about [TOPIC]. For this exercise, base your responses only on this document — not on your general knowledge. Tell me when you are drawing from the text and when you are inferring beyond it.

Stage 1 — Overview

Prompt
Summarize the main argument of this reading in 50 words.

Stage 2 — Dissect

Prompt
Identify the two most important paragraphs. Explain why they are central. Quote each one directly.

Stage 3 — Challenge

Use one of the two moves from Exercise 3 to push on something in the text.

Stage 4 — Cite

Prompt
Generate a reference for this reading in APA format. Include the author, date, title, and a DOI or ISBN link if available.
Prompt
Summarize what this reading is about, what I engaged with, and what I'm still uncertain about — based only on our exchange. Do not draw on outside knowledge.
Artifact: that summary. lp-ex8-[name].txt/artifacts.
Reflect
What does the model do well when confined to a single source? Where did it slip outside the text despite your instructions?
Transcript: lp-ex8-[name].txt/transcripts.
09 Second Reading and Comparison

The instructor provides a second reading. Upload it.

Prompt
I'm uploading a second reading about [TOPIC]. Base your responses only on this document for now.

Repeat the four stages from Exercise 8: overview, dissect, challenge, cite.

Then open the comparison:

Prompt
You may now draw on both readings. Where do they agree? Where do they contradict each other? What does each say that the other doesn't?
Prompt
Write a study reference comparing these two readings — their main arguments, their agreements and contradictions, and what together they suggest about [TOPIC].
Artifact: that comparison. lp-ex9-[name].txt/artifacts.
Reflect
Which reading made more sense to you? Did the model represent the differences honestly or flatten them?
Transcript: lp-ex9-[name].txt/transcripts.
10 Build Your Workbench

Open a new conversation. Upload artifacts from Exercises 2–9.

Prompt
I'm uploading a set of AI-generated artifacts from my conversations about [TOPIC]. What do they suggest about what I understand well, what remains unclear, and what questions I keep returning to?

Read the analysis.

Prompt
I have transcripts and artifacts from multiple AI conversations in a repository folder. I want to use this material in future conversations without re-reading everything. What would a good synthesis document contain? What are the risks of relying on one instead of the originals? Why might generating one still be useful?

Read carefully. The model is describing tradeoffs in AI-assisted knowledge compression.

Reflect
What did the model notice about your work that you didn't expect? What does your workbench reveal about how you've been thinking?
Prompt
Summarize your recommendations for organizing this workbench and your honest assessment of the tradeoffs of synthesis over source material.
Artifact: that summary. lp-ex10-[name].txt/artifacts.
Transcript: lp-ex10-[name].txt/transcripts.
Phase III — Independent Operation Exercises 11–15
Phase III · Exercises 11–15

Independent Operation

Set the destination, manage the route, document the trip.

From Exercise 11 forward, artifacts are documents — the kind that would be submitted as coursework or shared as outputs. Every artifact carries full citation:

Generated with AI assistance.

Student: [name]  |  Date: [date]  |  Model: [model name]

Transcript: [repository link]

11 Teach Yourself
Prompt
Create a learning sequence that teaches me [TOPIC] at four levels of depth. Use the same multiplier from Exercise 2 — or choose a new one. Each level must build on the last: by level four, the material should feel like something I already mostly know.

Work through each level. After each, tell the model what was new versus what was already understood. Push back if a level adds filler instead of depth.

Reflect
Did four levels feel like enough? Where did the progression break down?
Prompt
Compile the full sequence as a single clean document. Add full citation at the end.
Artifact: compiled sequence. lp-ex11-[name].txt/artifacts.
Transcript: lp-ex11-[name].txt/transcripts.

First Workbench Zip

After saving the artifact and transcript, zip the entire ai-literacy/ folder. Name the zip with today's date: ai-literacy-[name]-2026-03-01.zip. Save it to /zip.

This is a workbench snapshot — a timestamped record of where things stand at this point in the course. The synthesis document in Exercise 12 will be the entry point for future model sessions. The zip is the archive.

12 Synthesis Document

Upload artifacts from Exercises 4–11 and both readings from Exercises 8–9.

Prompt
I'm uploading my artifacts and readings from a course on [TOPIC]. Read them. Generate a synthesis document: an organized overview of what I've learned, what questions remain open, and what key ideas I should carry forward. Write it so it can be uploaded to any model as the opening context for a new conversation — portable, compressed, complete.

Read the synthesis. Push back on anything it misses or distorts.

Prompt
Explain what you did to generate this synthesis. What did you include and exclude, and why? What are the risks of relying on this document instead of the originals?

The model is describing its own compression process and naming the tradeoffs.

Reflect
Does the synthesis accurately represent what you learned? What did it miss? Would you trust it as the only context for a new conversation?
Prompt
Produce a final synthesis document incorporating my corrections, followed by your reflection on its limits. Add full citation.
Artifact: synthesis document with reflection. Title synthesis-[name].txt. Save to /synthesis. This is the entry point for future model sessions.
Transcript: lp-ex12-[name].txt/transcripts.
13 Did You Learn Your Own Slop?

Upload the synthesis document and artifacts from Exercises 6–12.

Choose a quiz format:

Option A
Based only on these artifacts — generated from my own conversations about [TOPIC] — create 20 multiple-choice questions that test whether I actually understand the material. Draw from what I wrote, not from general knowledge.
Option B
Based only on these artifacts, create 10 short-answer questions that test application and judgment, not recall.

Take the quiz in the chat. Write actual answers.

Prompt
Grade my answers honestly. Where did I demonstrate real understanding? Where did I reproduce language from my own artifacts without understanding it?
Reflect
What did you get wrong? Was it a gap in understanding or a gap in memory? Did you learn what you generated — or generate without learning?
Prompt
Write a document summarizing the quiz, my answers, your grading, and what this reveals about the relationship between generating and understanding. Add full citation.
Artifact: that document. Include actual answers — wrong ones included. lp-ex13-[name].txt/artifacts.
Transcript: lp-ex13-[name].txt/transcripts.
14 Go Public

Upload the synthesis document from Exercise 12.

Prompt
I'm uploading my synthesis document about [TOPIC]. For this exercise, base your responses only on this document. Do not draw on your general knowledge.

This is a test: can the synthesis document support a public-facing piece of writing without the model reaching beyond it? If it can, the synthesis is working. If the model struggles or slips outside it, that reveals what the synthesis is missing.

Prompt
Using only the uploaded synthesis document, write an introduction to [TOPIC] suitable for Wikipedia. Approximately 200–250 words, two sections, neutral tone, no first person, no hedging.

Read it as a reader. Would you trust it? What would you verify? Push back where needed.

Reflect
Where did the model stay within your synthesis? Where did it reach beyond it? What does that reveal about what you built?
Prompt
Final version, formatted cleanly. Add full citation.
Artifact: Wikipedia introduction. lp-ex14-[name].txt/artifacts.
Transcript: lp-ex14-[name].txt/transcripts.
15 Road Test — General AI License Exam

This is the exam.

Upload the synthesis document from Exercise 12. No scaffolding. The student decides the task, the format, the model, and the strategy.

Produce a substantive artifact about [TOPIC] — briefing, explainer, analysis, lesson plan, annotated summary, structured argument, or something else. Use the synthesis document as primary context. Instruct the model to work only from that document, or allow it to draw on general knowledge — but state the choice explicitly in the Prompter's Note.

The artifact must:

  1. Be the model's final response, shaped by the student's prompting
  2. Carry full citation
  3. End with a Prompter's Note — 3 sentences: what you tried, what didn't work, what you'd do differently

The transcript must:

  1. Be full and unedited
  2. Show the iteration, including what failed
  3. Be saved to /transcripts and linked inside the artifact
Reflect
Write in the chat before the final prompt. What did the permit teach you?
Artifact: lp-ex15-[name].txt/artifacts. Citation and Prompter's Note inside.
Transcript: lp-ex15-[name].txt/transcripts.

When both are accepted: the student holds a General AI License.

After Licensing General AI License

A General AI License holder may use generative AI independently for coursework, research, and professional tasks.

Every AI-assisted artifact carries full citation — name, date, model, and a live link to its transcript in the workbench repository.

The workbench — transcripts, artifacts, synthesis documents, author documents, and zip snapshots — is maintained continuously. It is not a course deliverable that ends at licensing. It is the ongoing record of AI-assisted work.

The workbench portfolio — demonstrating sustained, responsible use across courses and contexts — is the prerequisite for enrollment in the CDL Permit course.

Future Implementation Note

This scaffold runs in any environment with model access and a repository. A Brightspace or equivalent LMS container can house artifact submissions, link to repository folders, and track licensing completion. That integration is a next step, not a prerequisite.