Industries already have a safety language. We map AI into it.
Rather than invent a new AI risk vocabulary from scratch, AIC maps AI-assisted decisioning against the established safety and governance frameworks each industry already runs on — engineering's SIL ratings, banking's model risk management, medical software's safety classification. Each framework page states the positioning, the acceptable-use boundary, and the safety measures a subject is expected to demonstrate against — not AIC's internal assessment methodology.
Engineering / Process Industry
Where AI decisioning meets functional-safety engineering
Financial Services
Where AI-driven decisioning meets model risk management
Medical Devices & Health Software
Where AI/ML medical software meets safety classification
Reviewed 2026-09-01. These are the industries where we currently have enough genuine depth to publish a mapping — not an exhaustive list. We're researching further industries and will add them here once the same standard applies: real, publicly documented frameworks, honestly translated.
What these pages are, and aren't
Each framework page names the real, publicly documented standard it draws on, states AIC's positioning against it, and lists the safety measures a subject is expected to demonstrate — as outcomes. What's deliberately not published is AIC's internal scoring, evidence thresholds, or audit procedure — the same way a safety standard tells you what to demonstrate without telling you exactly how an individual auditor will judge whether you've demonstrated it well enough.