What this is

Tranquility is a tabletop demo for planning a massive lunar AI compute program—built to show how you structure decisions when the constraints are real (lift, cooling, governance, maintenance, logistics).

The scenario is lunar compute (one concept: thorium power on the Moon), but the method transfers to any infrastructure program that must survive scrutiny.

• Tranquility One: first 5 years — ~3.5 GW target.
• Tranquility Two: ~10 years later — ~35 GW target (trillion-class program; ~US$1.1T order-of-magnitude).

How to use it (1–2–3)

1. Download the ZIP — Download Tranquility Demo Pack (ZIP)
2. Paste the ZIP contents into the prompt of any AI tool you use (or upload the ZIP, if that tool supports file upload). Tell it: "Use this as the Tranquility demo pack."
3. Ask questions and iterate: constraints, tradeoffs, risks, and decision records.

Good starter questions to ask the AI

• "What are the top 10 constraints that kill this plan if ignored?"
• "What are 3 feasible architectures and why the others fail?"
• "Show me a risk register: top risks, owners, mitigations."
• "What governance model survives audit and partner scrutiny?"
• "What's the minimum viable Earthside program (HQ + lab) and why?"
• "What are the failure modes around lift, cooling, maintenance, and regolith?"
• "Draft the decision log for the first board meeting."