Status of Women Committee on Oct. 26th, 2026

[Emmanuella Lambropoulos, Liberal, Saint-Laurent, QC]

Thank you so much.

[Marilyn Gladu, Liberal, Sarnia—Lambton—Bkejwanong, ON]

Briefly, my question has to do with the education of people on AI. I'm pretty convinced that younger people are getting more education and training. They're using the tools more. I'm really worried, especially, about women who are out of the school system.

How can the government get at that population, which I think will be really left behind both in the way that AI is created and because of the fact that they're not getting trained?

[Indra Kubicek]

We realize that young people are using this in real time, having not had a lot of formal education. That's just the reality that these tools are available—

[The Chair, Conservative, Dominique Vien]

I'm terribly sorry, but the time allotted to the Liberal Party has expired.

Good morning, Ms. Larouche. You have six minutes.

[Andréanne Larouche, Bloc, Shefford, QC]

Thank you, Madam Chair.

I want to thank the witnesses for coming to speak with us today as part of our study. We're seeing more and more media coverage about the impact of AI. We're going to try to take a closer look at what this actually means for women.

Ms. Kubicek, briefly, do you want to add anything on the issue of education for individuals who are no longer in the school system?

I have another question for you, but I'll let you conclude your answer to my colleague's question, if you can do so in 15 seconds or less.

[Indra Kubicek]

Very quickly, I think we need to decentralize community-based learning spaces for people to get access to this free education outside of traditional education systems, to answer the last question.

[Andréanne Larouche, Bloc, Shefford, QC]

Perfect.

You said that Canada could be a leader, and that the issue of the disproportionate impact of artificial intelligence on certain groups in particular raised concerns at the UN.

I'm a member of the Canadian Group of the Inter-Parliamentary Union for my party, the Bloc Québécois. I have participated in a few missions with parliamentarians from around the world.

My first question might be a bit broad. I'm not sure how you're going to answer it. We can certainly talk about regulations here and what we can do here, but we realize that, when it comes to AI, we're going to need to take a much more global approach, because borders don't stop AI.

What's your response to that?

[Indra Kubicek]

For Canada to fully succeed in any of this, we need to work with our other middle-power partners, and that work is starting to happen already. The way we do that is by building programs with youth that can be shared internationally and developed with other young people in other countries.

For example, one of the biggest issues facing young people today is free access to chatbots, and they're using them in their social and personal lives. We just built a program with the Canadian Commission for UNESCO, working with young influencers to push this AI literacy out on social media. We took it to the UN in Paris so that we can also bring this in and work with other countries to get the word out on awareness through young people and with young people. That's one of the ways we're addressing it.

Absolutely, we can't do it alone, because a lot of the content, access, information and decisions young people are making through AI tools are not primarily and only in Canada.

[Andréanne Larouche, Bloc, Shefford, QC]

Thank you very much. That's very interesting.

I'm going to try to get to each witness.

I'll now turn to the two of you, Mr. Perry and Ms. Puodžiūnas. In your opening remarks, you spoke about the importance of bringing together the various stakeholders to discuss AI.

Do you want to tell us a bit more about that, Mr. Perry? Specifically, which stakeholders absolutely must be brought together? Furthermore, what issues should they focus on together, and in what context?

[Daniel Perry]

Thank you for the question.

Given the technical nature of the question, I will respond in English.

When it comes to stakeholders at the table, it's important to have industry there, especially the ones looking at this every day, because they understand what is working and what is not working. That is the benefit of CCI's membership. All of our members are founded here. They're mined and managed here. They are truly Canadian companies through and through. If we're talking about what will be impacting Canadians, who better to ask than the ones who are building their companies here—the employers in our communities, in your ridings and across Canada?

Canada was a founder of AI. We have a natural advantage, whether that's in our critical minerals or our education system. However, we've seen this story before. We were founders of Ozempic, and now it powers the Danish economy. The question we need to ask here is this: How can we be a part of it but also make sure we're keeping the companies here and growing them so they can be the BlackBerrys of the world and support innovation here?

[Andréanne Larouche, Bloc, Shefford, QC]

Ms. Puodžiūnas, I'll continue my questions with you.

You said eliminating bias was impossible. So, we're starting from the premise that AI has gender bias. We know that, in Montreal, there have been demonstrations showing that AI has racial bias. We've seen clear examples of these biases. Could you tell us a little more about the exact nature of gender bias in AI? Why did you say it would be impossible to eliminate?

Maybe I misunderstood what you said. Could you clarify this for me?

[Skaidra Puodžiūnas]

I said that in the spirit of illustrating the importance, again, of the fact that these systems are being built by humans, and humans, of course, have bias. Therefore, the systems we build are going to be as biased as the data we power them with.

Again, there's such an opportunity to ensure that we are building these systems with the localized perspective of domestic innovators—of the Canadians who will ultimately be using them—in order to ensure that they have a Canadian perspective and a Canadian lens on them.

That was the point I really wanted to illustrate in my commentary on that.

[Andréanne Larouche, Bloc, Shefford, QC]

Perfect.

I have 30 seconds remaining, so I will quickly conclude with you, Ms. Flood.

You said that responsible and inclusive intelligence requires investment. In just a few seconds—we can come back to this during the next round of questions—could you tell us what such investments should prioritize?

[Amy Flood]

I recommend prioritizing a national sex-, gender- and diversity-disaggregated data strategy.

[The Chair, Conservative, Dominique Vien]

That's all you have time for. Perfect timing. Thank you very much.

We will now begin our second round of questions.

Ms. Borrelli, you have the floor.

[Kathy Borrelli, Conservative, Windsor—Tecumseh—Lakeshore, ON]

Thank you, Madam Chair.

Can you tell the committee more about why the Council of Canadian Innovators favours a standards-based approach to AI governance, and how standards can help reduce bias and discrimination in AI systems?

[Daniel Perry]

That is a great question, and it is one we talk about in our AI consultation paper.

The reality is that legislation moves slowly. We have seen privacy legislation attempted three times now, and online harms legislation twice. In that time, AI has changed. Even as I'm talking now, AI will be a little different by the time I finish this sentence. That's the challenge of legislation. It does not move at the pace of industry.

That's where standards can help. Standards provide that baseline, but they also signal to companies where the government's intentions are. This is at the heart of why we're calling for standards. They are the best way to govern this. As we're building the plane, they would give us the runway and the guidance to move forward on it in ways legislation would not.

[Kathy Borrelli, Conservative, Windsor—Tecumseh—Lakeshore, ON]

How can the federal government help Canadian AI companies grow while making sure AI works for all Canadians?

[Daniel Perry]

Procurement is the easiest way to do that. Invest in our companies through government contracts. That is how we're able to grow them, and that's also how the government can be part of the conversation. Set out what we need so we can build it specifically for the government.

As my colleagues have mentioned regarding health data, for instance, we don't provide that to our innovators. We provide it to some research institutions but not to business. When we are building our AI datasets in the health care system, they're being built on U.S. or European data. Their health concerns are very different from our health concerns. We need good information and good datasets. That is one way to do it. Provide that. That's something the federal government can do, and we would encourage them to.

That was in one of our recommendations to the minister's AI task force as well.

[Kathy Borrelli, Conservative, Windsor—Tecumseh—Lakeshore, ON]

Thank you.

To what extent are AI developers responsible for identifying, preventing and addressing biases and discrimination in AI systems during the programming stages and development process?

[Daniel Perry]

I think my colleague might be a little bit better suited for that.

Our view is that the developers should be responsible for explaining how it was made and how it was put together. The deployers, the companies, are responsible for discussing how it is being used, what data is being used on it, how it's being trained and whether a human is involved.

[Kathy Borrelli, Conservative, Windsor—Tecumseh—Lakeshore, ON]

Indra, do you have any comments in that regard?

[Indra Kubicek]

It's not exactly my background, but I would agree with that. There needs to be an onus on the developer to develop it responsibly so that it's going to work ethically in the way people want. On how it's deployed now that it's in the company's hands and how they're training their users and their employees, they also have to own some of that responsibility if that's where the error or the flaws occur.

[Kathy Borrelli, Conservative, Windsor—Tecumseh—Lakeshore, ON]

Are there examples of countries that are leading in the development and implementation of AI technologies while maintaining the effective regulatory frameworks that ensure AI systems are safe and do not discriminate against marginalized communities?

Skaidra, do you have...?

[Skaidra Puodžiūnas]

From our perspective, we've been seeing a lot of inspiration from the U.K. government as well as Australia, particularly looking at it through a Commonwealth lens. There is a real move to incorporate more of an MOU approach. Instead of adding yet another regulatory body or another regulatory legislative piece that, again, will take years to pass, there's a real horizontality that we're seeing in some of these models. From an innovator perspective, that's a really important approach, because innovation moves horizontally. It's thinking about different contexts and different use cases.

If our legislative vehicles can look at it from that perspective, connect those dots and perhaps be better wayfinders for innovators navigating our Westminster model, approaches like looking at the Australian and U.K. governments and some of the approaches they've been taking would be a good model for that.

Of course, I am not being very specific, so I will give some concrete examples to the committee after this, but these are two countries that we're looking at for inspiration.