After Session 1 | ~20-25 minutes
You just completed your first session. You built something real with AI, and you noticed things — what worked, what didn't, what surprised you. This expedition is about processing that experience and pushing into new territory. One reflection, one new build, one honest evaluation, one plan forward.
Pull up the AI conversation from your Session 1 build. Look at it with fresh eyes.
Answer these three questions:
- What worked? What did AI do well? What part of the output was actually useful?
- What felt uncertain? Where did you hesitate, or feel like you didn't know what to ask for?
- What would you do differently? If you ran that exact task again right now, what would you change about how you approached it?
If you don't have the conversation saved, work from memory. What you remember is enough.
Choose a different task type from what you did in Session 1. Not a variation on the same thing — genuinely different terrain.
- If your S1 task was writing (email, report, memo) → try analysis, planning, summarization, or research
- If your S1 task was analysis or research → try drafting, creative problem-solving, or a communication task
- Not sure what counts as different? Ask yourself: "Was S1 about creating something, or about understanding something?" Do the other one.
Requirements:
- Real work. Not a practice problem. Something from your actual job this week.
- Build a first draft with AI. Give it your best shot with what you know so far.
- Save or copy the key output — you'll need it for the next step.
Look at what you just produced. Apply the quality check:
- Would you use this as-is? Or does it need work before it goes anywhere?
- What's missing? What would make it actually good enough to send?
- Circle one specific thing that's wrong. Not a vague feeling — one concrete flaw. A wrong assumption, a missing audience consideration, a tone mismatch, a factual gap. Name it.
Finding the flaw is the skill. You don't need to fix it right now.
Create your Delegation Plan — the three tasks from your real work that are most suitable for AI assistance.
Rank them by difficulty:
| Difficulty | What it means |
|---|---|
| 1 | Routine, well-defined. You've seen AI handle it. Feels safe. |
| 2 | Somewhat complex or unfamiliar. Reasonably confident, but it's a stretch. |
| 3 | Genuinely uncertain. You'd be surprised if AI handled it well — but it's worth trying. |
For each task, write one sentence about why you ranked it where you did.
Compile your work from the four steps into this format. This is your artifact — the thing you share and bring to Session 2.
EXPEDITION JOURNAL — [Your Name]
Date:
S1 REFLECTION
- What worked:
- What felt uncertain:
- What I'd do differently:
NEW BUILD — [Task name and type]
- Task I attempted:
- How I approached it (1-2 sentences):
- Key output (paste or describe briefly):
- Quality check — one thing that was wrong:
DELEGATION PLAN — Top 3 AI-Suitable Tasks
| # | Task | Difficulty (1-3) | Why I ranked it here |
|---|------|-------------------|----------------------|
| 1 | | | |
| 2 | | | |
| 3 | | | |
Send your Expedition Journal Entry to your trail partner before Session 2. Even a rough draft counts — they're counting on having something to respond to.
This journal opens Session 2. Your trail partner will read it. The facilitator will use your "what would you do differently" reflections to set up the next concept. What you wrote in Step 1 is not just processing — it's the data that makes the next session land.
If you can only do one thing, do Step 2 — Build. The cross-domain build is the highest-value step. Coming to Session 2 having tried AI on a second task type — even without the reflection and plan — gives you the experience the session builds on.
RANGE dimensions touched: Reach (cross-domain build), Execution Fidelity (evaluate your output), Navigation (plan your delegation strategy)