AINODIS helps organisations deploy AI responsibly - starting with how people actually work, not how we assume they do. Observation first. Policy last.
Some will cause harm they didn't anticipate. A handful will get it right - and the difference is almost never the technology. It's whether someone took the time to observe, map, and govern before deploying. That's what AINODIS does.
- §01 / The premiseShadow the people doing the work. Map what's actually decided, escalated, and quietly overridden - before any policy is drafted.
Translate observations into a concrete risk & decision map: where AI touches consequential decisions, where accountability sits, where it doesn't.
Policies, review gates, escalation paths - written as product, not policy. Tested against how people actually behave under pressure.
Ongoing retainer for leadership teams who need a partner in the room - reviewing pilots, pressure-testing decisions, and keeping governance lived, not laminated.
Full-time embedded leadership while you build the permanent function. Set up the operating model, hire your successor, leave the structure standing.
A two-week intensive: contextual inquiry with a specific team, a risk map you can act on, and a workshop that turns it into concrete next moves.
Most AI policies I've read describe a company that doesn't exist. AINODIS starts from the one that does.
Head of Risk, European financial services group
A 28-page working paper: the premise, the three moves, the failure modes to watch for, and the artefacts produced at each stage. Free to read. Cite it if it's useful.
# ainodis/framework.yaml premise: "Governance follows observation, not precedes it." ─────────────────────────────────────────────── moves: - id: 01 name: observe artefact: field-notes.md - id: 02 name: map artefact: risk-surface.svg - id: 03 name: govern artefact: operating-model.pdf ─────────────────────────────────────────────── anti-patterns: - "Policy without fieldwork" - "Governance as marketing" - "Owning the risk register, not the risk"
.png)
A 45-minute working session, no slides. Bring the system you're about to deploy - or the one that's already quietly making decisions. We'll map the first risk surface together.
A 45-minute working session, no slides. Bring the system you're about to deploy - or the one that's already quietly making decisions. We'll map the first risk surface together.