What 200+ FinOps practitioners learned about AI commitments.
The AI Infrastructure Commitment Checklist: 25 questions drawn from 200+ practitioner talks across financial services, technology, retail, media and manufacturing, so you can defend the number to finance. Drop your email and I’ll send it to you.
What’s inside
Six areas to cover before you commit. Most recur across 30+ practitioner talks. Supply, the least covered, adds operations research and industry data, plus what 12 years of running cloud capacity at Google taught me.
Demand
Base, high and shock scenarios, and what each one does to your commitment.
Supply
Chip, memory, power and lead-time constraints, shortage playbooks, and supply as a growth enabler.
Hardware and models
How new chips, models and services could change your cost per unit mid-term.
Contract terms
Ramp, true-down, transfer and exit flexibility, and the utilization floor you must hit.
Efficiency
What utilization and optimization gains you are assuming, and who owns them.
ROI and sign-off
The end-to-end investment view finance will want before approving.