Quebec’s Law 25 and AI: what’s changing for SMBs

“We use a Canadian server, so we’re compliant.” It’s a phrase you often hear when discussing AI and Law 25. It’s also one of the most wrong.

Quebec’s Law 25 isn’t limited to data residency. It changes how you must manage every AI project that touches personal information. Here are the four blind spots.

1. Purposes and consent

Personal information may only be used for the purposes for which it was collected. Using customer emails or call transcripts for a new purpose, such as training or evaluating an AI model, generally requires new consent, subject to exceptions set out in the law. That consent must be manifest, free, informed and given for specific purposes. A generic checkbox at sign-up isn’t enough.

2. Privacy impact assessments (PIA)

The law requires a PIA for any project to acquire, develop or overhaul an information system or electronic service delivery system involving personal information. Most AI projects qualify. A PIA is also required before communicating personal information outside Quebec, which is the case with many cloud AI providers. It’s not a formality: it’s a documented analysis of risks and the measures taken.

3. Automated decisions

If you use personal information to make a decision based exclusively on automated processing (screening applicants, approving credit, prioritizing files), you must inform the person no later than when you inform them of the decision. On request, you must also tell them what information was used and the main factors that led to the decision, and let them submit observations to a staff member who can review it. Keeping a human in the loop isn’t just good practice.

4. The role of the privacy officer

By default, the person in charge of the protection of personal information is the person with the highest authority in the company; the role can be delegated in writing. Their title and contact information must be published on the company’s website. They should be involved early in AI projects, especially for the PIA.

What to do

Before launching an AI project that touches personal information: map the data processed and its purposes, conduct a PIA, check where the AI provider processes the data, plan for destroying or anonymizing information once its purposes are fulfilled, document automated decisions and involve the privacy officer. It’s less heavy than it sounds when done at the diagnostic phase. It’s unmanageable when added at the end.

This article is for information only and is not legal advice. For your specific situation, consult a lawyer or Quebec’s Commission d’accès à l’information.

Have an AI project?

Let's talk. 30 minutes to understand your context and identify the real opportunities.

Start a conversation