The Weekly Brief — Issue 18

Ottawa is going further than grants and research funding in its effort to build an AI industry at home.

The federal government plans to invest up to $1.6 billion for an equity stake in Cohere, the Toronto-founded AI company. The investment will happen in stages, alongside commitments from Cohere to add Canadian jobs and computing infrastructure.

The bigger shift is what the government says the investment represents: AI capability is increasingly being treated as part of Canada's economic security and sovereignty.

Plus: major Canadian banks are restricting personal AI agents from online banking, BDC has launched two AI initiatives, including one to help small businesses adopt the technology, and Canada's privacy regulators are calling for stronger safeguards when AI is used in health care.

Here is the week.

What happened

The federal government has committed to an equity investment of up to $1.6 billion in Cohere.

Ottawa says its investment will be made in stages, with regular reporting on Cohere's commitments in Canada.

Those commitments include at least 1,000 new Canadian technical jobs by 2030, at least $1.7 billion in additional computing infrastructure in Canada by 2030, keeping Cohere incorporated and headquartered here, and investing in Canadian research partnerships and opportunities for students and recent graduates.

The precise commercial terms are not public. We do not know Canada's eventual ownership percentage or how much will be invested at each stage. AI Minister Evan Solomon told BetaKit the stages are tied to milestones, with jobs and revenue targets key. He did not share the terms.

Cohere is also raising money from private investors. Reuters reports that its financing round is targeting $7 billion at a valuation of roughly $36 billion, in Canadian dollars. Only an initial close has occurred. Cohere says that covers most of the target, but it has not disclosed the amount.

Why it matters

Canada has spent decades funding AI research. The Cohere deal moves federal support closer to the commercial market.

Ottawa is not simply supporting research or offering an incentive to build here. It plans to own part of a commercial AI company and share in its potential gains — and its risks.

The government also places the deal inside a broader argument about sovereignty. Announcing the deal the same day Ottawa released a new national security strategy, Prime Minister Mark Carney said sovereignty now requires secure access to technologies including semiconductors, critical minerals, clean energy technologies and AI.

That does not make Cohere's technology "sovereign" in every practical sense. A German retail group is leading the financing, Nvidia is also investing, and Cohere is merging with Germany's Aleph Alpha. AI systems still depend on chips, cloud infrastructure, energy, data and global supply chains.

But it shows how Ottawa is defining the problem: Canada wants more of the companies, computing capacity and expertise behind AI to remain anchored in Canada. It is also looking to allies: Canada joined Europe's EuroHPC supercomputing network on October 30, which could widen access for Canadian researchers and businesses.

The distinction to remember: Canada has committed up to $1.6 billion to become a Cohere shareholder. That is not the same as $1.6 billion already being spent, and Cohere's promised jobs and computing investment are commitments for the years ahead — not results already delivered.

Your move

Watch the conditions, not only the headline number.

The useful questions now are how much public money is actually invested, which milestones Cohere must meet before later stages proceed, whether the required reporting will be made public and whether the promised jobs and computing infrastructure materialize by 2030. Thursday's federal budget is the next place to look for any added detail on Ottawa's AI plans.

Several of Canada's largest banks are restricting third-party AI agents from directly accessing authenticated banking services.

RBC says it is blocking personal AI agents because of risks involving credentials, fraud and consumer harm. TD says it is blocking third-party agents from authenticated banking information, while BMO says AI agents cannot access its digital properties or execute transactions. Scotiabank says it will limit third-party agents, while CIBC says it does not support them directly opening or accessing accounts.

Why it matters: AI agents can do more than answer questions. With enough permission, they can interact with other software and take actions on a user's behalf.

That becomes much more consequential when the connected system contains money.

The banks are not rejecting AI altogether. They are drawing a boundary around third-party agents using customer credentials while questions about authorization, security and responsibility remain unresolved.

For users, the immediate rule is simple: do not give an AI tool your banking password because it promises to make managing your money easier. RBC's digital banking guarantee excludes losses if you share your passwords, so check your own bank's terms.

The Business Development Bank of Canada launched Mission AI and AI Studio on November 2.

Mission AI will identify common SME problems, match businesses with Canadian AI products and test those tools in real companies. AI Studio will develop and test AI inside BDC, including tools intended to simplify processes and improve access to financing.

BDC says 69% of Canadian SMEs in its latest survey are AI novices, meaning they use AI very rarely or not at all, while only 5% have deeply integrated AI.

Why it matters: most Canadian businesses are not frontier AI companies. For them, the question is whether an affordable tool can reduce administrative work, help prepare a quote or make an everyday process faster.

Mission AI is an attempt to close that adoption gap by testing what works before asking more businesses to follow. The results still have to be demonstrated.

Canada's federal, provincial and territorial privacy regulators publicly released a joint resolution last week calling for stronger safeguards and governance around AI used in health care and health research.

The resolution was adopted September 15 and released publicly October 29. It asks governments to review, amend or adopt laws and regulations, consult regulators and the public, and legally require effective AI-governance frameworks.

Why it matters: health data is among the most sensitive information Canadians generate, and AI tools can influence decisions involving diagnosis, research, administration and care.

But this is not a new national health-AI law. The resolution sets out what Canada's privacy regulators believe governments, health organizations, researchers and AI providers should do. Existing privacy laws still apply, but the new obligations it proposes would require governments to act.

An equity investment means buying ownership in a company.

That is different from a grant, which generally provides money for an approved purpose without giving the government an ownership stake.

With Cohere, Ottawa says it is making a commercial investment. If the company becomes more valuable, Canada's stake could become more valuable too. If it struggles, the investment carries risk.

We do not yet know enough about the commercial terms to calculate Canada's ownership percentage or potential return.

The important point is simpler: Canada plans to become one of Cohere's owners, not just one of its funders.

When an AI assistant offers to connect to another service, check what it is asking permission to access.

Ask four questions: What information can it see? What can it change? When does it need your approval? Can you see and stop what it is doing?

This matters most for accounts involving money, sensitive personal information or the ability to send, delete or change something on your behalf.

An AI agent can be useful without receiving unlimited authority.

A calm, practical briefing on the AI developments that matter in Canada—delivered every Tuesday.

That's the week. See you next Tuesday.

— Padge T.

AI Brief Canada is an editorially independent publication published by Groundshift Advisory Inc. and written by Padge T.

Groundshift Advisory helps Canadian organizations adopt AI responsibly. Sponsored, affiliated, or commercially connected content will always be clearly disclosed.

Issue disclosure: Nothing in this issue is sponsored, affiliated, paid, or commercially connected.

AI use: AI tools supported research, source discovery, outlining and drafting for this issue. All material was reviewed, edited and fact-checked by Padge T. before publication. Read our AI Use Disclosure.

Not affiliated with the Government of Canada.

Corrections or feedback: We welcome corrections and reader feedback. Email [email protected] or review our Editorial Policy.