Thank you, Chair and honourable members of the committee. I appreciate the opportunity to speak to you all today.
My name is Cam Linke. I'm the CEO of AMII, the Alberta Machine Intelligence Institute, one of Canada's three national centres of excellence in artificial intelligence.
Our mission encompasses fundamental research, applied innovation and the diffusion of technology, specifically artificial intelligence, across Canada, which I believe provides us with a unique vantage point on the committee's study.
To start, Canada should take a lot of pride in its scientific leadership in artificial intelligence. Researchers such as Richard Sutton, Geoffrey Hinton and Yoshua Bengio, our three Turing award laureates—one of whom did win a Nobel Prize—laid the foundations of modern artificial intelligence and positioned Canada as a global leader in AI research excellence. In fact, if you look, you'll see that nearly all of the core AI advances that are powering the global AI revolution today can be traced back to researchers at Canada's three AI institutes.
That said, investment in AI research and development by Canadian industry has been relatively weak. Today I'm going to focus on three challenges that we see in the Canadian economy and three things that we found helpful in overcoming them.
At AMII, we've now worked with hundreds of companies, from small start-ups and SMEs to large firms, and we consistently see three structural barriers across Canada that limit private sector engagement.
The first, just to be clear, is the structure of Canada's economy, and I think we need to be realistic about this.
In Canada, we have a small number of large firms, which often operate in protected markets, and a very large number of SMEs. These large, structurally protected firms often lack the competitive pressure to pursue ambitious R and D. Further, SMEs, which make up the vast majority of Canadian businesses, frequently want to innovate but lack the capital or expertise to do so, and since a relatively low number of graduate student researchers are working within Canadian companies, many of those companies lack the internal capacity to identify and execute meaningful projects.
The second barrier is around our funding architecture. Canada tends to be a funding ecosystem that is heavy on grants but light on risk capital. We offer a number of programs, but they're often dominated by small, fragmented grants that require significant administrative overhead to achieve impact. Grant funding in and of itself isn't a bad thing; it's great to be able to make targeted investments. That said, compared to risk capital, often in the form of venture capital, grant funding leads to smaller and less impactful applied research being done in Canadian companies.
A third barrier is the absence of foundational research infrastructure. Companies can't invest in R and D if the necessary infrastructure isn't in place. Access to compute, specialized facilities, secure environments and data pipelines that connect universities to industry are essential. Without these, firms face an impossible choice—build the infrastructure themselves, which is often cost-prohibitive, or take the research elsewhere. As an example, many AI advances that began here in Canada found their commercial impact outside of Canada, where the necessary computing resources were available.
At AMII, we've seen these three challenges first-hand and worked with our partners to tackle them. I'd like to highlight three ways that we've seen success.
The first is in early-stage support. At AMII, our advanced technology program collaborates with companies to identify research opportunities that drive business value. Then we work with them very closely to make sure that the research project succeeds. For SMEs in Canada, this early expertise helps de-risk the project without requiring significant upfront investment and enables them to source the talent necessary to progress in their projects. It should be noted that typically this company engagement is made without grant support, which I'm saying as a positive thing. This is where companies have seen the value in these engagements and have seen the groundwork to be able to invest.
Next, let's consider the availability of computing infrastructure.
A great deal has been said about the computing requirements needed for success in AI. Programs like the pan-Canadian AI compute environment, recent announcements of compute in the federal budget and partnerships with Canadian companies such as Denvr Dataworks have provided capacity for research industry collaborations to take place. These have led to larger collaborations that drive impact in domains such as health care and energy, where the scale of these projects wouldn't have been possible without the available infrastructure.
It should be noted also that outside the area of AI, we've seen investments like these drive clusters of impact. For example, many quantum and chip start-ups collaborate with researchers at the University of Alberta due to the government's investment in the university's nanotechnology fabrication facilities.
Finally, we have found that breaking down the so-called language barriers has been vital in creating opportunities for industry to invest in research. At AMII, we talk a lot about “bilingual researchers”, researchers who speak both the languages of a specific domain like energy, biology or health care, as well as artificial intelligence. As part of our recent hiring of new principal investigators to our group, we targeted researchers who would provide this dual expertise. Bringing in researchers who deeply understand the domains of application have broken down the barriers that at times can exist between academia and industry. We expect this to continue to drive increasing competitive advantage for Canadian companies in the coming years.
To summarize, targeted research investments, cross-domain researchers and providing early project hand-holding and talent can lower the barrier to entry for research investment by Canadian industry and increase research and development overall.
In order for Canada to lead both in AI and other fields in the coming years, we need Canadian industry as well as governments to invest—
