Is your AI ready for production — or still a demo?

Ten questions, drawn from the Forward-Deployed Playbook, that separate an impressive demo from a system that survives real data, cost, and a security review. Answer honestly — nothing is sent anywhere; this runs entirely in your browser.

What the ten questions actually measure

The questions aren't a quiz about AI knowledge — they're the checklist I run in the first week of a real engagement, compressed. They probe the five things that decide whether an AI system survives contact with production: whether success criteria exist in writing (a task, a bar, and a named sign-off, rather than "it looks good"); whether you can evaluate pass/fail against that bar on your real data, repeatably; whether you know your cost and latency at actual volume, not demo volume; whether the system would survive a security review — prompt injection, data exposure, what the agent is permitted to do; and whether a named person on your team can own and operate it after launch.

Most teams score lower than they expect, and that's normal — the gap between a convincing demo and dependable production is where almost every AI initiative stalls. A low score doesn't mean the project is in trouble; it means the remaining work is now visible and sequenceable. The Forward-Deployed Playbook walks through closing each gap in order, and the fixed-scope AI Production Readiness Audit is the done-with-you version: same dimensions, applied to your actual system, with a prioritized plan at the end.

Your answers never leave this page — there's no form, no tracking of responses, and nothing to submit. If the result surprises you in either direction, that's usually worth a conversation.