Clinics, dental offices, tax preparers, lawyers and financial advisors handle some of the most sensitive information their clients have. AI tools can save these practices real time, but pasting client details into a public chatbot can break privacy rules and client trust.
In December 2023, Canada’s federal, provincial and territorial privacy regulators, including BC’s, published principles for using generative AI. For a small practice, they come down to a few practical rules.
1. Know what data goes where
Before using any AI tool, find out where the data is stored, whether it is used to train the vendor’s models and who can access it. Business versions of many tools offer stronger protections than free versions.
2. Share as little as possible
The regulators call for limiting the sharing of personal, sensitive or confidential information. Remove names, health numbers, SINs and account numbers from anything you send to an AI tool unless the tool is approved for that data.
3. Be open with clients
Tell clients if and how you use AI in your work, in your privacy notice and in plain language when it matters. Transparency builds trust.
4. A professional checks every output
AI drafts can be wrong in confident language. Any advice, diagnosis, tax position or document that reaches a client is reviewed and owned by a qualified person.
5. Write it down
A one-page AI policy covers which tools are approved, what data may go into them, who reviews outputs and how staff report a mistake. It protects the practice and makes training new staff easier.
Safe places to start
- Drafting general client education material, such as aftercare instructions or a tax deadline checklist
- Summarizing internal meeting notes without client identifiers
- Rewriting your website and policies in plain language
How WIZAD can help: we help you choose approved tools, write a one-page AI policy and set up the first safe use cases with your team.
Source: Office of the Information and Privacy Commissioner for BC, Canadian privacy regulators launch principles for the responsible development and use of generative AI (December 2023).
