President of the Treasury Board, Anita Anand says guidelines are needed. Photo: File.
The Canadian Federal Government has introduced guidelines for employees who want to use artificial intelligence tools like ChatGPT on the job.
President of the Treasury Board, Anita Anand said the aim was to ensure the technology was being used responsibly.
“We will also be monitoring the way AI is being used to guard against potential problems like bias or discrimination,” Ms Anand said.
“As a racialised woman myself, I am very conscious about the potential for bias to creep into decision-making.”
She said the guidelines complemented the existing directives to Government Departments on artificial intelligence and would be updated as needed.
“While there are currently no penalties for violating the new guidelines, they are based on existing legislation such as the Privacy Act that could trigger a penalty,” Ms Anand said.
While the Treasury Board’s guidelines for generative AI recommend that Federal institutions explore ways to use these tools, it also warns of risks, including cyber-security threats, bias, violations of privacy and inaccurate information.
The guidelines define generative AI as technology that “produces content such as text, audio, code, videos and images” for things like chatbots, e-mails, briefing notes, research or programing.
The guidelines recommend caution when using AI for things like public communications on social media or automating “assessments, recommendations or decisions about clients”.
If a Department uses generative AI to respond to a citizen, answer questions via a chatbot, create a document or make a decision, it should be transparent about using the technology, the guidelines state.
Departments should “identify content that has been produced using generative AI, notify users that they are interacting with an AI tool, document decisions and be able to provide explanations if tools are used to support decision-making,” they say.
Ottawa, 13 September 2023
