Human Resources Committee on Nov. 1st, 2023

[Tracy Gray, Conservative, Kelowna—Lake Country, BC]

Thank you, Mr. Chair, and thank you to all of the witnesses for being here.

My first questions will be for the Council of Canadian Innovators.

In your opening remarks, you talked about low productivity being “not normal for advanced economies”.

I'm wondering if you can explain that or go into a bit more detail on what you meant by that.

[Laurent Carbonneau]

It's a very big question.

If you look at our standard of living, our incomes and a lot of these things, you would think that this is a very productive economy, but if you look under the hood, it's less productive than you would expect. There are a lot of reasons for it, which are historical and have to do with industrial structure....

That's what I mean, fundamentally.

[Tracy Gray, Conservative, Kelowna—Lake Country, BC]

How does that relate to Canadian citizens? What does that mean for Canadians and also for the business climate here?

[Laurent Carbonneau]

I'll pivot back to something Mr. Roberts said, which was about this not being the be-all and end-all. I think that's entirely fair. It really does matter where the money goes. It's all about how much wealth we are generating as a country, and then where the wealth goes is very much the stuff of politics. That's all fair; I don't disagree with that.

To that point, the impact of this is how wide our horizon is as a country. If we're not a very productive economy compared to our neighbours, then it just forecloses options for us in the future. It forecloses options in terms of how we deal with problems like climate change, reconciliation and poverty. It just limits the real resources that we have at our disposal for dealing with those social problems and for enjoying a high standard of living.

[Tracy Gray, Conservative, Kelowna—Lake Country, BC]

That's great. Thank you.

I'll go over to the OECD representative on that same topic—

[The Chair, Liberal, Bobby Morrissey]

She has gone.

[Tracy Gray, Conservative, Kelowna—Lake Country, BC]

Oh, has she gone? Okay. I knew that at some point she was going to remove herself.

I'll go back to the Council of Canadian Innovators. Where are we trending right now in terms of productivity?

[Laurent Carbonneau]

I mentioned this in my opening statement. From 2018 Q2 to now, there have been little wobbles, but we're actually a little below where we were in 2018 Q2. That's pretty worrying, isn't it? You expect to see a low but reasonably consistent level of productivity growth. Obviously, these last five years have been odd in lots of different ways, but other countries have not had that experience.

It will be very worrying, I think, if we see that trend continue. We hope it doesn't, but I think there are some structural reasons to think that it might continue to be the case.

[Tracy Gray, Conservative, Kelowna—Lake Country, BC]

How do we compare with other countries? Where are we in scale, I guess you could say. How do we compare with our largest trading partner, the U.S.?

[Laurent Carbonneau]

I can speak to the U.S. part. We are well behind the U.S. There's a pretty big gap between output per hour.

On the others, I honestly would have to get back to you. I have a couple of comparisons in my statement there, saying that we're right around where Italy, Turkey and Spain are, but I don't remember the exact top 10 or however many.

[Tracy Gray, Conservative, Kelowna—Lake Country, BC]

On the topic of the U.S., in “A Roadmap for Responsible AI Leadership in Canada”, you referenced that the U.S. was doing work on this. They came out this week with AI rules in the U.S. I'm wondering if you could comment on whether you've had a chance to go through those yet. Do you have any thoughts on those?

[Laurent Carbonneau]

Yes, I took a quick look. I would say it's a pretty impressive suite of measures. I think, broadly speaking, it's very good. It goes further than what we have as a legislative framework currently, which is frankly what you'd expect. I think the current AI bill is saying that this is sort of how we will frame rules, and then let's create those rules.

The executive order is really a very different approach. It's what they can do through executive action. There are some really interesting measures in there around standards. There are some really interesting measures around the audit of large language models that might have security implications.

Overall, I think it's a really impressive package of work. I would have to look a bit more into the details of how that's going to shake out, but I think our first impressions were reasonably positive.

[Tracy Gray, Conservative, Kelowna—Lake Country, BC]

I have one more quick question here. I'm almost out of time.

Without going into the weeds at this point, do you think there's some benefit for Canada to harmonize with countries like the U.S., and maybe other countries—to really look at harmonization? Would that be something that Canada should be considering?

[Laurent Carbonneau]

Absolutely, yes: I think that's the short answer. As I mentioned earlier, when it comes to scaling AI companies, Canada's not a huge global market. If companies are able to compete globally from day one because our regulatory system is aligned with those of the EU and the U.S., that is a very big advantage.

[The Chair, Liberal, Bobby Morrissey]

Thank you, Ms. Gray.

We'll go to Mr. Kusmierczyk for five minutes.

[Irek Kusmierczyk, Liberal, Windsor—Tecumseh, ON]

Thank you, Mr. Chair.

This is an absolutely fascinating and timely discussion. Today President Joe Biden tweeted about his executive order on AI. I know that tomorrow the U.K. will be hosting an international summit on AI safety. Again, this is a very, very timely discussion. I want to thank my colleague, Michael Coteau, for bringing this study forward to the committee.

Mr. Roberts, I was very interested to read the report that was published by the CLC and the Pembina Institute and that just came out in September, “A Sustainable Jobs Blueprint”. It talks about how Canada can build a framework of supports that can help workers transition to a zero-emission economy. It's a great read. I highly recommend it. It really dovetails nicely with the legislation that's in front of Parliament right now, Bill C-50, which we brought forward.

Can you talk about some of the parallels or lessons from that conversation or from that process and how they might apply to this conversation about the challenges, whether it's AI or whether it's automation or digitization, and about what lessons and parallels we can draw from that?

[Chris Roberts]

The labour movement, for a long time, has championed the notion of just transition to a sustainable economy, with the idea being that workers have to see their own future in the new economy, otherwise they won't support moving to a net-zero arrangement. The idea was that workers have to participate in the process. They have to be involved, and their interests have to be included in the transition.

The same goes for AI, I would argue. If workers are involved, if they see a pathway to a prosperous future for themselves and for their family, they will be involved. There will be a social licence to participate.

That also means we have to think about digital skills, about digitalization of work as a similar transition to the transformations in work that will be required to move to a sustainable economy. That means investing in labour adjustment mechanisms, ensuring that training is much more equitably distributed than it is today and that continuous and lifelong learning opportunities are available to enable mid-career workers, for instance, to adjust, and the like.

Your analogy is very appropriate.

[Irek Kusmierczyk, Liberal, Windsor—Tecumseh, ON]

Do you have specific recommendations in terms of how the voices of labour and the voices of workers could be included in that conversation?

Our government has put forward over $400 million towards the pan-Canadian AI strategy. We have centres set up across the country looking at this issue.

How do we get workers and labour—their voice—into that conversation?

[Chris Roberts]

Bill C-50 and the sustainable jobs framework contemplate a partnership council that will bring workers, industry and other interests together to address the skills and labour market programming needs to make that transition.

We can similarly fashion institutions that bring workers and their organizations into the process. There is an AI advisory council currently. Unfortunately, it's constituted exclusively by industry and academics. There are no civil society, labour or human rights advocacy organizations, etc.

That would be a starting point—to create a body that can identify skills needs and programs that are required, collaboratively, and instill some of that social licence that's required. Then, I think having sectoral tables as well that identify industry and sector-specific needs in response to AI development and the digitalization of work would take us a long way down the road.

[Irek Kusmierczyk, Liberal, Windsor—Tecumseh, ON]

Absolutely.

I'm going to share, Chair, the bulk of my time with my colleague Mr. Fragiskatos.

[The Chair, Liberal, Bobby Morrissey]

You have 30 seconds.

[Peter Fragiskatos, Liberal, London North Centre, ON]

We'll make it 30 seconds.

Mr. Carbonneau, you mentioned productivity in general terms at the outset of your presentation. A few days ago I met with a representative from the Canadian Home Builders' Association, who brought up permitting.

We know that permits for housing at the municipal level often take a very long time to issue. Without getting into the reasons for that—because there are many—they mentioned AI as a solution to this. They pointed out that AI can approve a permit, according to all the guidelines that municipalities would want approved, in a matter of just a few minutes—or a few hours if it's not a few minutes.

What do you think about that as a solution to the housing challenges that are in front of the country right now?

[The Chair, Liberal, Bobby Morrissey]

Give a short answer.

[Laurent Carbonneau]

I'll zoom out a touch, while keeping it short. I would say there is a broad opportunity for the public sector in general to use AI in a lot more of its operations.

For that to be the case, I think the way we approach procurement will probably have to change a bit to allow for a more back-and-forth dialogue between innovators and folks who want to buy innovation in the public sector.

We have structures that don't really enable that right now. Innovation is essentially an iterative process. That doesn't really flow with how things work right now.

I think that would be fantastic. I think there are obviously lots of use cases for AI in the public sector, housing included, but we'd have to look at those structures and see how we can make them work better.

[The Chair, Liberal, Bobby Morrissey]

Thank you, Mr. Fragiskatos.

Ms. Chabot, you have two and a half minutes.