Where to go from here
What this is
Chapter 10 ends the book on a deliberate note. “AI is not magic. It gets things wrong. It can be confidently, fluently, convincingly wrong… Everything in this book that worked, worked because I brought real problems, real constraints, and real judgement to the conversation. AI made my thinking better. It didn’t replace it.” The tools are available — most of them free. The skills are the ones you already have: describing what you want, pushing back, keeping going when the first answer isn’t right.
This exercise is the bridge from reading to using. By the end of the next twenty to twenty-five minutes you’ll have had your first serious AI conversation about something that actually matters to you — not a test, not an experiment, but a real problem, decision, or goal you’ve been carrying around. Then you’ll set up what happens for the next seven days, so this isn’t a one-off.
What you’ll need
- 20–25 minutes
- One real thing you’ve been thinking about while reading this book — a decision you’ve been avoiding, a project you’ve been meaning to start, a skill you’ve been telling yourself you’ll get to
- Your usual AI tool open in a fresh conversation
Pick your one thing
The right thing for this exercise has three properties.
- It matters to you. Not the most urgent thing, the biggest thing. Something you care enough about to describe honestly. If you pick something you don’t actually care about, you’ll generate a polite version of yourself for the conversation and get a polite version of advice back
- You’ve been carrying it. Long enough that you have a sense of why it’s hard. Long enough that there are reasons you’ve been putting it off. The conversation works on what you bring; what you’ve been carrying is what you bring
- You don’t already know the answer. If you’ve decided and you’re just looking for AI to validate the decision, this is the wrong exercise. Pick something where you genuinely don’t know what to do — or where you suspect what you think you know might be wrong
Don’t overthink the choice. The right thing is the one you keep coming back to.
Step 1 — describe what’s on your mind, and ask AI to interview you first
The opening move from Chapter 4. Not “help me with X”. A situation, honestly described, plus an explicit ask for AI to interview you before suggesting anything.
I want to talk about [what’s on your mind — a decision, a problem, a goal, a project you’ve been putting off]. Here’s my situation: [describe it honestly in a few sentences — the current situation, and why you are stuck].
I don’t want generic advice. Interview me first — ask me the questions I haven’t asked myself, including the uncomfortable ones. Then give me your first take on what you’re hearing.
Answer the interview honestly. Including the questions you don’t have a great answer for. Especially those. The answer “I don’t know” is more useful than a polished guess — it shows you and AI both where the actual gap is.
Once you’ve finished the interview, read AI’s first take the way Chapter 5 taught you to read AI’s first response: not as a finished answer, but as a starting point. Push back on anything that feels generic.
Step 2 — find what you’re underweighting, then commit to a next step
Before you finish, run one more move. This is the Chapter 6 principle (let AI critique itself) applied to your conversation rather than to a document.
Before we finish, two questions.
Based on what I’ve told you in this conversation, what am I most likely underweighting in this decision or situation? Give me specific evidence from what I’ve said, not a general observation.
And: what’s the single most useful thing I could do in the next seven days to move this forward?
The first question often surfaces something you’d noticed but hadn’t said out loud. The second question gives you a concrete next step — not a plan, a thing to do.
What comes after this conversation
You’ve had one serious conversation. The book’s argument is that the value compounds when you keep going. Three resources on this site are designed for what comes next.
The 30-day practice path (/book/practice). Four weeks of daily practice. The first seven days mirror the book’s “first week with AI” table; the rest builds from there.
The prompt library (/book/prompts). Sixteen starter prompts across seven categories. Mirrors Appendix B.
The chapter exercises (/book/exercises). If you want to practise a specific move again — building something, starting a coaching relationship, getting a second opinion, comparing tools — the per-chapter exercises are designed to be revisited.
If you only do one of these things, do the seven-day plan. One conversation a day for a week is enough to build a habit. Anything less and what you’ve read in the book stays as knowledge, not practice.
What you’ll notice
The interview usually surfaces something you hadn’t articulated to yourself. People often go into this exercise thinking they know what they’re carrying. AI’s questions — especially the ones that hit something uncomfortable — sharpen the problem in ways that the reader’s own framing wouldn’t have done.
The single most useful thing you could do in the next seven days will sound small. “Have one specific conversation with X this week” or “Write the first paragraph of the thing you’ve been avoiding” or “Make the call you’ve been postponing”. That’s the right scale. If AI gives you a project plan, you didn’t push hard enough for the single most useful thing.
The book ends with a line worth carrying out of this exercise: “It has never been easier to go from having an idea to making it real.” That doesn’t mean it’s easy. It means the barrier that used to be there — the gap between knowing what you wanted and being able to make it — has come down enough that the bottleneck is now what you bring to the conversation.