Q&A: This think tank rated the Liberal AI strategy poorly for sovereignty. Here’s why

Ottawa’s recent artificial intelligence (AI) strategy provoked commentary on a variety of subjects, including jobs, education, and investment, but the Canadian Shield Institute scored the policy poorly in an area that has been a point of emphasis for the federal government — sovereignty. The think tank issued a report giving the Liberal strategy a six out of 10 for sovereignty, citing governance concerns and a lack of clarity around the technology’s forecasted economic impacts. In an exclusive interview, Matthew da Mota, the institute’s research director for emerging technology, unpacked the report card and discussed why the government’s discussion around sovereignty and AI is lacking honesty. Your organization gave the federal government’s AI strategy a six out of 10 regarding sovereignty. What went into that rating? The first thing I would stress is that we’re not totally negative about it. It shows a lot of positive signals. But, to have a strategy, you need to be strategic. That means picking priorities in a little bit more of a concerted way. They receive so many positions from so many different organizations, and they’re trying to really satisfy as many people as possible, and that’s bound to lead to some issues. The main places where it lost marks was the lack of clarity on how we’re going to be supporting Canadian IP and ensuring we’re holding that here, how we’re going to be prioritizing Canadian companies, and where the sovereignty argument is going. A lot of the positions we’ve seen so far reflected in the strategy imply a significant amount of cooperation with U.S. hyperscalers. What is that actually going to look like? How are we going to preserve sovereignty and security in that process? Those are some of the core elements. Is your major concern that these questions weren’t addressed at all? Or, are you looking for specific solutions that weren’t included? Some of them are not addressed. There’s mention of IP, but there’s not a lot of detail. In other cases, you might be open to giving the benefit of the doubt, but this strategy has existed since 2017. As an example, we have the three ‘godfathers of AI,’ all of whom were very famous researchers at Canadian universities. They invented some of the core IP or came up with the core research that led to huge advancements in AI funded by Canadian research funding. But that was never commercialized. We never really made an effort to hold on to those ideas, turn those into companies and ideas and products here, and the result is companies like Google got those ideas for free. Now, we couldn’t have assumed this jump forward would happen, but these are structural issues with how we deal with research funding. In other areas, like education, it makes sense not to give a ton of detail yet. That’s going to take time and collaboration with provinces to figure out. The other big thing was the focus on jobs. We gave it no points for job creation because the job creation numbers they gave are very speculative. It didn’t address the huge question that most people are concerned about, which is that the promise of AI seems to be able to fire a lot of people. I think they can figure these things out. It’s not a hopeless strategy, and I think there are a lot of positive elements, but there needs to be more clarity on this, especially when they’ve been working on it for a year. Prime Minister Mark Carney (Nepean, Ont.) has been very vocal about the importance of sovereignty over the last year. It sounds like you believe the government’s strategy hasn’t lived up to its rhetoric? I think sovereignty and AI are very complicated. No country, not even the U.S., has full control over the entire value chain of AI. We have to be very specific and deliberate about understanding where we can actually control part of the value chain and expand it. That’s important, but then also making sure we have controls in place for areas where we can’t have full sovereignty or control. From what I’ve seen, the intention is to build real sovereignty and I think sometimes how that ends up rolling out is complicated and difficult, but we haven’t seen as much movement forward as we would have liked. For all the gaps you’ve outlined, what would you like to see the government do? Are you looking for a revised strategy? Legislative changes? I think it’s really key that we’re getting foundational legislation on privacy and data.* That’s really important. The online harms proposal is a good start, though it doesn’t fully address the structural issues, which are issues of antitrust. I’d like to see a more strategic approach to data protection, but also mobilization. We have some of the best data in the world across our institutions, so we need to figure out how to provide that access to institutions, researchers, and SMEs, so they can thrive. I think those would be the next steps. It doesn’t require a new strategy, but they’re really going to be important in building a foundation. Can you paint a picture for me? If the government embraces all these recommendations, what would Canada’s relationship with AI look like? It would look like a more honest appraisal of our relationship to American companies, and a clear admission of where we’re going to have to accept ceding some sovereignty, and not pretending it’s sovereignty because of the partnership we’re making. We hear this term called ‘picking champions,’ but we need real clarity on what that means because I think it needs to mean picking core areas of the AI supply chain that we’re really going to invest and try to lead on. Take Cohere, that’s a company that’s continuing to do great work, but we need to pick the zones we want to build up. That would also mean a really comprehensive strategy about how we’re dealing with data centres and compute capacity. Those would be the key things, and then just continuing to promote and support SMEs, but not just with funding. I think they’ve done very well getting funding to companies, but they really need to think about the next stages. How do we grow those companies? How can we provide markets for them? How does the government become a customer in a way that enables and attracts more capital? If there’s so much money coming from the sector, and we’re investing so much in that, we really need to think about how that can enable SMEs. This interview has been edited for length and clarity. dlegree@thewirereport.ca *Editor’s note: This interview took place before the government introduced its privacy bill earlier this week.

Q&A: This think tank rated the Liberal AI strategy poorly for sovereignty. Here’s why

Regulatory |
By Davis Legree
|June 19, 2026

Unlock all the Canadian telecom, broadcasting and digital media news you need.

FREE TRIAL

SUBSCRIBE

Inside The Wire Report

Information

Hill Times Publishing

Get in Touch With Us

Q&A: This think tank rated the Liberal AI strategy poorly for sovereignty. Here’s why

Regulatory | By Davis Legree |June 19, 2026

Unlock all the Canadian telecom, broadcasting and digital media news you need.

FREE TRIAL

SUBSCRIBE

Various digital policy, telecom bills stuck in Parliament as MPs break for summer

Cabinet declines to overturn CRTC Iristel decision

Rogers warned about illegal fees by CRTC

Related Stories

Liberals to introduce law guaranteeing ‘right to privacy’ on Monday

Young people polled on AI and deliver recommendations to government

Fraser hints deepfake regulations could extend into privacy law as opposition MPs say AI strategy lacks protection details

‘Stealing is not innovation’: SOCAN CEO calls on feds to stand against text and data mining exemption for AI developers

Reuse & Permissions

Inside The Wire Report

Information

Hill Times Publishing

Get in Touch With Us

Regulatory |
By Davis Legree
|June 19, 2026