Governance

Clear AI rules, understood by everyone

Define your internal AI usage rules, share them, then measure whether they're truly understood.

A policy no one understands protects no one

The Québec government's guide and Law 25 expect organisations to frame AI use with clear rules. But a policy filed on an intranet, read by no one, offers no real protection. Aptia closes the loop everyone forgets: not only defining and sharing your rules, but verifying they're understood.

How it works

How it works

1

Start from a template

An assistant proposes an AI usage policy adapted to the Québec context.

2

Adapt to your reality

Adjust the rules to your sector, your tools and your risk level.

3

Share it

Send the policy to your teams in a few clicks, in a clear version.

4

Measure understanding

A short check confirms the policy is read and understood, not just sent.

What's inside

What an AI usage policy covers

Concrete rules, tied to the competencies the diagnostic assesses.

Data & confidentiality

What can, and cannot, go into an AI tool, in the sense of Law 25.

Human verification

The duty to review and validate, the antidote to hallucinations.

Approved tools & shadow AI

Which tools are allowed, and how to avoid unframed use.

Ethics & bias

The precautions for decisions with high human impact.

The difference:
measuring understanding

It's what no one else does. Because the policy is tied to the same competencies as the diagnostic, you don't just share it: you know who understood it, and where grey areas remain to clarify. A policy becomes a living measure, not a dead document.

Clear rules, understood by all

Start by measuring where your teams stand: the free diagnostic reveals the gaps your policies will need to close.

No credit card · No commitment · Confidential data