The federal government is asking Canadians for their views on artificial intelligence — specifically, whether people should be told when they are dealing with an AI system, and who is responsible when an AI tool gives them wrong information. The consultation, called "Have your say: Advancing AI transparency in Canada," is posted on the Innovation, Science and Economic Development Canada website. It includes a survey and is open until September 23. (The Globe and Mail; ISED)

A discussion paper released with the consultation lays out five topics for the public to consider. The paper deals with questions about consumer protection and legal responsibility that current laws were not written to handle.

One of the biggest questions is who should be held responsible if an AI chatbot gives someone bad information. The paper notes that Canadians are already using AI agents to shop online, send emails, and book meetings. It points to a real example: a Toronto man got an offer from a car dealership's AI chatbot, named Quinn, to buy back his vehicle for $27,163. The offer was wrong, and the chatbot did not tell the man he was talking to a computer. This is the kind of problem the government wants to address. (The Globe and Mail)

The paper also raises a newer concern. AI agents are now talking to other AI agents — for example, one company's shopping bot might deal with another company's sales bot — without a human involved. When these systems miscommunicate, it can be hard to figure out who is at fault and what rules should apply.

The paper notes that the European Union and several U.S. states already require companies to tell people when they are interacting with an AI system. Canada has not yet put similar rules in place at the federal level. The consultation is meant to help the government decide whether and how to do that, though it says no decision has been made yet.

This consultation follows the launch of Canada's federal AI strategy in June. (The Globe and Mail; ISED) Together, the strategy and the consultation show the government is starting to move from general promises about AI toward specific rules on transparency and accountability.

The broader context is about timing. The EU and some U.S. states already have transparency rules, and Canadian companies that operate across borders may already be following them in practice. A consultation that runs until late September means any new rules would likely come after Parliament resumes sitting and after the AI market has changed further. The paper's mention of AI agents talking to other AI agents acknowledges that the technology is moving fast. Whether new laws can keep up is a question the September 23 deadline does not answer.

For anyone affected by these rules, the key thing to know is the scope. This consultation is not about AI safety in general or about research funding. It is specifically about transparency: when people must be told they are dealing with AI, who is responsible when the system makes a mistake, and what rules should apply when AI tools act on their own on a user's behalf. The five topics in the discussion paper, the survey, and the September 23 deadline define the window in which Canadians can have a say.