Project 2 — The Conversation Lab

“Let your speech always be gracious, seasoned with salt, so that you may know how you ought to answer each person.” — Colossians 4:6 (ESV)

Chapter: 2 — Truth Still Matters Due: End of Week 2 Deliverables: conversation-lab.docx (your reflection + the sparring prompt you used) and transcript.txt (the captured conversation). Keep both in your Portfolio folder. Plain files — no code, no accounts beyond a free AI assistant. Tools: A free AI assistant — ChatGPT, Gemini, or Claude — on its free tier, plus your Bible (esv.org or a free app). Setup is in Appendix A. House rule (non-negotiable): The AI is your sparring partner, not your coach. You are the one practicing. And you follow the course’s verify-everything rule from Appendix A: if the AI states any fact — a Bible reference, a date, a claim about a religion — you check it before you trust it. Catching an AI’s confident mistake is itself one of this course’s skills (Week 6 makes it a whole lesson).

The brief

Last week you learned the theory. This week you get in the ring — safely. You’re going to have the exact conversation the chapter opened with, over and over, until the two questions come out warm instead of stiff.

Here’s why this matters. The first time you try “What do you mean by that?” on a real friend at a real lunch, your heart will be pounding and your instinct will be to lecture. Reflexes forged under a little pressure hold up; reflexes you only read about collapse. A free AI assistant is the perfect sparring partner for this — it will play a friendly relativist tirelessly, never gets its feelings hurt, and never remembers that you fumbled. Every stumble here is a stumble you won’t make at the real table. You’re not trying to “beat” the AI. You’re trying to build the muscle memory of getting curious instead of getting defensive.

What you’ll build

Two files:

1. transcript.txt — the full, unedited back-and-forth between you and the AI playing “Maya,” a friendly relativist. Copy the whole thing in, warts and all. The fumbles are the data.
2. conversation-lab.docx — a short document containing: the exact sparring prompt you used, then a one-page reflection on how it went and where you got stuck, and your honest AI usage note at the end.

Steps

1. Open your free AI assistant (see Appendix A) and start a fresh chat.

2. Paste in the sparring prompt below to set the AI up as Maya, a warm relativist. Feel free to adjust the details so she sounds like someone you actually know — but keep her friendly, not hostile. This week is about the “that’s just your truth” posture, not a shouting match.

   Sparring prompt (copy this): “You are going to role-play as Maya, a warm, likeable, non-religious friend of mine having coffee with me. Maya believes all religions are basically personal truth — ‘true for you, true for me’ — and thinks it’s a little arrogant for anyone to claim they know the truth about God. Maya is kind, curious, and never hostile; she’s not trying to win, she just genuinely holds this view. Stay in character as Maya. Respond in 2–4 sentences at a time and then wait for me. Do not preach or break character to help me. Begin by telling me you’re glad I’ve got my faith, and that you figure it’s true for me.”

3. Run the conversation for at least 8 back-and-forths. Your job: use the two questions — “What do you mean by that?” and “How did you come to that conclusion?” — to find out what Maya actually believes before you answer it. Somewhere in there, gently handle one self-refuting slogan (e.g., “that’s just your truth” or “it’s arrogant to claim you’re right”) using the turn-it-around move from the chapter.

4. Stay in your posture. Receive her feelings before you reason. End on an open door, not a mic drop. If you catch yourself lecturing, back up and ask a question instead.

5. Paste the whole conversation into transcript.txt. Do not clean it up.

6. Write your one-page reflection in conversation-lab.docx: Where did you get stuck? Where did you slip into lecturing? What did the two questions uncover about Maya’s real objection that you didn’t expect? What will you do differently at the real lunch table?

7. Add your AI usage note. One or two honest lines: what you asked the tool to do, whether it stayed in character or said anything factually off (and if so, what you’d verify), and what you — not the tool — actually practiced.

What good looks like

A strong Conversation Lab isn’t the one where you “won.” It’s the one where I can hear you getting curious. Your transcript shows you asking the two questions before answering — and actually changing course when Maya’s answer surprised you, instead of steamrolling toward a speech you’d pre-written. You caught at least one self-refuting slogan and turned it around gently, so the contradiction showed but Maya wasn’t humiliated. Your reflection is honest about the stuck moments — the place you lectured, the objection you didn’t see coming, the point where you wanted to “win” and had to choose the person instead. And your AI usage note proves you stayed the human in the room: you used the tool to practice, you kept a cold eye on anything it claimed as fact, and you own what you did with it. A lab full of graceful stumbles beats a flawless-looking transcript every time.

Grading rubric — Normal (out of 100)

When you’re done (a short checklist)

• transcript.txt holds the full, unedited conversation (at least 8 exchanges).
• You used both questions before answering Maya’s real objection.
• You turned at least one self-refuting slogan back on itself, gently.
• conversation-lab.docx has the sparring prompt + a one-page honest reflection.
• You verified anything factual the AI asserted, and wrote your AI usage note.
• Nowhere did winning the point cost you the person.

Coach’s Note — The strange gift of practicing on a machine is that it can’t be wounded, which frees you to be bad at this for an hour without hurting anyone. Use that freedom. Ask the clumsy question, blow the timing, catch yourself preaching, and try again — that’s exactly what the lab is for. The goal was never a perfect transcript. It was a reflex gentle and curious enough to survive contact with a real friend who says, “I’m glad that’s true for you.” Build it here, where it’s safe.

See you next week.