Trust

Our AI usage standard

Practical AI only works if people can trust it. These are the rules we follow.

How do we use AI in our own work?

We use AI tools to speed up research, first drafts and analysis. A person on our team reviews, edits and approves everything before it reaches a client or goes live. We don't publish unreviewed AI content on client websites.

What rules apply to the AI systems we build?

Disclosure: customers are told when they're talking to an AI assistant, and when calls are recorded.
Human handoff: a person is always reachable, and sensitive or uncertain conversations are escalated.
Approved knowledge only: the AI answers from information the client has approved, with clear limits on what it may promise.
Data minimization: we collect only what the workflow needs and store it in accounts the client owns.
Consent: marketing texts and emails follow Canada's Anti-Spam Legislation (CASL).
Review: conversations are logged and reviewed regularly, and mistakes are fixed in the instructions.

Which privacy laws do we follow?

We design AI deployments to support clients' obligations under BC's Personal Information Protection Act (PIPA), the federal Personal Information Protection and Electronic Documents Act (PIPEDA) where it applies, and CASL. We document which vendors process data and where. This page is not legal advice; clients in regulated fields should confirm requirements with their own advisors.

Do we use client data to train AI models?

No. We choose vendor settings that don't use client data for model training wherever the vendor offers that option, and we tell clients where it doesn't.