Vector Institute

And one of their co-founders had been a student of Jeff Hinton's.

And so they basically started the lab or the firm, I should say, around Jeff's lab at the University of Toronto, co-founded the Vector Institute in Toronto, all these significant institutions in the world of AI.

And they were interested in starting a presence here in the UK and Europe more broadly.

So it just felt like a pretty amazing opportunity to get to work with these people who had been deeply immersed in this ecosystem for as long as they had.

So the questions that seem to come up all the time are what role should it play in an advisory practice? Where are the opportunities? What are the risks? And most importantly, how can advisors embrace these capabilities while preserving the client trust and fulfilling their obligations that are central to their practices? That is exactly the set of questions that we're here to explore today.

Joining me is Cameron Schuler, Chief Commercialization Officer and Vice President, Industry Innovation at the Vector Institute.

which is one of Canada's leading artificial intelligence research organizations.

Cameron has spent years helping bridge the gap between breakthrough innovation and practical adoption, working with industry, research and government leaders to turn emerging technology into real-world impact.

So we're not just using tools.

For example, we're partnered with the Vector Institute at the University of Toronto.

It's Professor Hinton's lab.

He's the Nobel Prize winner in AI.

especially when there are words containing specific sounds and political videos online.

Shanaya Raza, an applied machine learning scientist for responsible AI at the Vector Institute for Artificial Intelligence in Toronto, shares other red flags voters can watch for to spot AI-generated misinformation ahead of Election Day.

Be cautious about materials appearing in the final days before an election.

So deep fakes can be particularly effective during elections because they often involve emotional or controversial claims, and people's feeds can create echo chambers where the same misleading content is repeatedly shared and amplified.

Roles with high AI exposure also carry newly added tasks calling on judgment, empathy, and creativity at a rate roughly two and a half times greater than less exposed roles.

A separate report titled Artificial Intelligence Talent in Canada, produced by the Conference Board of Canada for the Vector Institute alongside the Future Skills Centre and released in October 2024, found that employer demand for core AI skills climbed 37% over the five years from 2018 to 2023.

Demand for peripheral AI skills covering software development, design, and cloud computing contracted by 46.4% over that same period.

The report attributes that decline to automation tools taking on a growing share of those tasks.

Among specific competencies, operating AI systems expanded by 48%, while overseeing machine learning projects climbed by 60%.

The report flagged AI governance as the area where Canada's existing strengths create the greatest opportunity for further development.

Melissa Judd, Vector Institute's Vice President of Research Operations and Academic Partnerships, said in the October 2024 announcement that bridging technical expertise with business and domain-specific applications alongside expanded governance capacity would be the path forward.

Those findings present a direct challenge to short-form AI training programs.

Raquel is pioneering the field of physical AI, starting with the most scalable and safest self-driving solution.

Raquel is a full professor in the Department of Computer Science at the University of Toronto and co-founded the Vector Institute for AI.

Raquel, how are you doing today?

I'm doing fantastic and really a pleasure to be here with you, Harry, today.

They put together a project team for a health research grant.

They now have dozens of psychiatrists that are part of this process with us, AI experts, and we're going to go through the Vector Institute to be commercialized.

Aonest will have AI on it.

That also turns into its own problem set because the second you put AI onto a software system, all of a sudden people get to scrape all the data, so we lose data sovereignty.