Industry and Technology Committee on June 1st, 2026

[Dominique O'Rourke, Liberal, Guelph, ON]

We are hearing in this conversation and seeing in this stuff we're reading that “Data is the new oil”. We're hearing about this race.

Mr. Gallant, I want to ask you about the importance of sovereign launch capacity here in Canada and the importance of having sovereign satellites and ownership of the data that can be used for everything from agriculture to forest fires, climate change, etc. Can you tell us why that's important and what some of those applications will look like?

Then, if we have time left, I'd like to come back to Mr. Sookman and hear his recommendations.

[Brian Gallant]

The last question your colleague asked also dives into what I would say about this. One thing that countries are doing is that they're picking the capabilities they want to be able to develop. When it comes to government playing a role as a customer or as investment, it can also be that they're maybe the only game in town, or a very big percentage, as we saw with SpaceX at the beginning.

As my colleague mentioned, if you are going to put restrictions on who can invest in their companies because you want to keep IP, that makes a lot of sense, but then you need to recognize that they need some support domestically to be able to do that and to ensure they have the capital they need to grow. You can't say, “Don't go get investors from another country,” but also “We have no money here for you in this domestically to be able to do so.”

Then, for launch specifically, I think.... This is not to speak on behalf of the government, but it seems like that is one that they indeed picked. They said, “This is something that we are going to build: sovereign capability.” I think the importance of it is that if ever we have some type of disruption, like a war, a conflict—and this literally has already happened to our sector—or we see trade conflicts, which has already happened to our country, we will not have to worry about our ability to get things up into space when we need to.

Given, to your point, that it is critical infrastructure, we need that capability domestically, and the government has to play a big role in funding it and making it happen, which it is.

[Dominique O'Rourke, Liberal, Guelph, ON]

I'll go to my last question. I'm hearing a bit of contradictory testimony. I'm hearing that we need some controls over a number of things.

Mr. Sookman, you seem to be saying that we need the least amount of regulation possible, except maybe later when the systems are more advanced. You said that you have a few recommendations. I would like to hear what those are.

[Barry Sookman]

I want to address the particular problem that was mentioned about the protection of data, because that is a critical problem. What's often lost is the distinction between different kinds of models.

With the generative AI model, there seems to be a lot of push-back related to anybody owning that output. In fact, there are cases around the world that have said that outputs from generative AI models are not protected by copyright. Inventions that arise from generative AI models are likely also not protected. What that misses in that discussion is the difference between different kinds of models and what they do.

When you're talking about the data that's being collected in this manner, there is a much different public policy for protecting it. There was a case a number of years ago involving geophysical data and the protection of information that was harvested through seismic exploration. That was found to be protectable. There's no conceptual difference between protecting that seismic data that wasn't created using AI and protecting some data that's protected from deep learning models, which is what we're talking about, in order to ensure that it can be proprietary. Right now, the law is unclear in Canada, and it's very likely that if that got tested to the extent that the overwhelming thing that shaped the data was the AI, it would not be protected.

There are things that can be done to create a sui generis right, similar to what the European Union did with respect to databases. There are ways to do it, and that issue is really critical when you have investments like this.

[The Vice-Chair, Conservative, Raquel Dancho]

Thank you very much.

We've had excellent testimony today. I appreciate you all being here. Thank you very much.

Colleagues, we'll briefly suspend while we switch over to the next panel.

Thank you very much.

[The Vice-Chair, Conservative, Raquel Dancho]

I call this meeting back to order.

We will resume our study on opportunities, risks and regulation of AI in Canada's strategic industries.

Welcome to the new panel of witnesses.

You have my apologies. We're starting a bit tardy today. We did have votes that delayed our commencement of the committee. Thank you very much for your patience.

Everyone is appearing remotely today. I believe they've completed their required connection tests in advance of our meeting, so everything should work. If it isn't working, we will pause, fix any issue and resume your time accordingly.

We are welcoming four witnesses this afternoon. From the Coalition for the Diversity of Cultural Expressions, we have Marie-Julie Desrochers, executive director. From ControlAI, we have Samuel Buteau, consulting program officer. From Copibec, we have Christian Laforce executive director. From Sustainable AI Group, we have Sasha Luccioni, co-founder and chief scientific officer.

As you each know, you'll have five minutes for your opening remarks. I will cut you off at five minutes. My apologies in advance, if I have to do so.

We will begin with Madame Desrochers.

The floor is yours for five minutes.

[Marie-Julie Desrochers, Executive Director, Coalition for the Diversity of Cultural Expressions]

Thank you very much, Madam Chair.

My name is Marie‑Julie Desrochers and I am the executive director of the Coalition for the Diversity of Cultural Expressions. For over 25 years, the organization has been carrying the collective voice of Canada's cultural sector in all its diversity. We bring together more than 50 organizations, representing 350,000 creators, as well as 3,000 cultural enterprises across multiple disciplines.

I am joined by Christian Laforce of Copibec, a member of the organization's executive board. Last week, your committee heard from the Society of Composers, Authors and Music Publishers of Canada, which is also a member of the organization.

Generative artificial intelligence is becoming an unavoidable force and is profoundly transforming the cultural ecosystem. Some embrace it and integrate it into their daily work, while others proceed more cautiously or reject it altogether. Different levels of adoption and comfort coexist. However, beyond these differences, one point of consensus remains: the need for proper regulation so that this technology serves human creativity, rather than the other way around.

The work of the Coalition for the Diversity of Cultural Expressions is grounded in the 2005 UNESCO Convention on the Protection and Promotion of the Diversity of Cultural Expressions. Canada played a decisive role in the development and adoption of this convention. It was the first country to ratify it. Twenty years later, this commitment remains essential. Canada is a recognized leader in the implementation of the convention, a role that must continue in the era of generative artificial intelligence.

In this regard, we have in fact received encouraging signals this year from the government of Canada. In February, at UNESCO, the Minister of Culture and Canadian Identity, Mr. Marc Miller, supported the addition of a protocol to the convention. This marked a significant step toward the adoption of a new binding legal instrument aimed at complementing, strengthening and enriching the objectives of the convention in the digital environment, particularly in response to the growing challenges posed by artificial generative intelligence systems.

During the national summit on artificial intelligence and culture held in Banff this past March, Minister Miller also stated that he had no intention of amending the Copyright Act and noted the importance of ensuring that rights holders are compensated when their works are used. These are encouraging statements, but they still lack clarity and firmness. We therefore strongly hope that the national artificial intelligence strategy will finally provide the necessary guarantees, at a time when discussions on the North American framework and the place of cultural policies within it take on particular importance in the context of the review of the Canada—United States—Mexico Agreement.

The Coalition for the Diversity of Cultural Expressions' positions are part of a broad international consensus and are well within reach for the government of Canada. They can be summarized in three words forming the acronym ART: authorization, remuneration and transparency.

Let us start with authorization. The Copyright Act requires industrial users to obtain authorization from rights holders prior to any use of their works, and this must remain unchanged. To be clear, there should be no new consent mechanism introduced, but rather a reaffirmation that Canada operates in an opt-in system, and a clear statement that Canada does not intend to modify this principle, for instance, by introducing an exception for text and data mining.

Next, with respect to remuneration, we are at the dawn of a new licencing market for rights holders, based on voluntary, individual or collective licencing schemes. This market should not be disrupted or undermined.

Finally, when it comes to transparency, developers and deployers of generative artificial intelligence systems should be required to disclose the training data used and to document their sources. In addition, content generated by artificial intelligence systems should be clearly identified as such when made public.

This past February, the Coalition for the Diversity of Cultural Expressions organized a conference entitled Valuing Human Creativity in the Age of AI. Speakers from France, Australia and the United Kingdom came to support the Canadian cultural sector's positions and highlight developments in their respective jurisdictions.

Sebastian Cuttill, head of parliamentary and legislative affairs at the News Media Association in the United Kingdom, where the government recently stepped back from its intention to introduce a text and data mining exception, aptly stated that respecting copyright is not about

protecting the vested interests of a particular sector or maintaining an outdated legal system. It is simply about applying a basic principle of resource management: when someone wants to use your assets, you must ensure that the value created remains within your economy.

We fully endorse this view. Genuine innovation cannot be built on the appropriation of such a valuable resource as human creativity and our intellectual property, but on the recognition of its value in every sense of the word.

Thank you for your attention.

[The Vice-Chair, Conservative, Raquel Dancho]

Thank you very much, Ms. Desrochers.

We will go now to Mr. Laforce.

[Christian Laforce, Executive Director, Copibec]

Good afternoon. My name is Christian Laforce and I am the executive director of Copibec.

Copibec is a Quebec-based non-profit organization specializing in copyright management. It was founded in 1997 and represents over 30,000 authors and more than 1,400 publishing houses. Copibec facilitates legal access to a vast repertoire of works protected by the Copyright Act through licencing.

Copibec has entered into agreements with 34 foreign companies to include publications from these countries in its repertoire. This year, we have distributed over $13 million to rights holders, for a total of $300 million since Copibec's inception.

Over the past few years, Copibec has been involved in a number of files on artificial intelligence, including the federal consultation on generative artificial intelligence in 2024 and work on Bill C‑27.

More recently, Copibec appeared before the Standing Committee on Canadian Heritage for its study on the effects of artificial intelligence on the creative industries. Our observations can be found in the report that the committee presented to the House of Commons last April.

Generative artificial intelligence is evolving at record speed and is profoundly reshaping cultural industries. Generative artificial intelligence companies receive substantial investments from governments, and they have quickly drawn on the extensive works produced by Canadian authors, and have done so while disregarding the Copyright Act. I would remind the committee that this legislation gives rights holders exclusive rights, and as such, the use of any protected works to train generative artificial intelligence systems requires authorization, particularly when it comes to text and data mining.

Compliance with the Copyright Act must be ensured in the context of artificial generative intelligence and any artificial intelligence regulation must include this requirement. In this regard, Copibec supports the principles of authorization, remuneration and transparency, or ART, which have been brought forward by the Coalition for the Diversity of Cultural Expressions, of which we are a member, and which Marie‑Julie Desrochers spoke to earlier.

In addition, Copibec recommends that the Copyright Act should not have any exception for text and data mining. We hope Minister Solomon will retain these recommendations in his national artificial intelligence strategy, which will reportedly be unveiled soon.

In this regard, a number of comparable countries, including European Union member countries, the United Kingdom and Australia, have affirmed the need to comply with copyright when training artificial intelligence models. Canada should follow their lead.

In May 2025, Copibec signed an exclusive partnership in Canada with Botscorner, which has leading-edge technology to help publishers identify bots and give them concrete tools to negotiate, block or enter into licencing agreements to use their content. The use of this tool has shown us that some bots scrape content, despite the use of exclusion files, such as robots.txt files, which create rules that are applicable to these types of automated activities.

Similarly, we have launched a study to review the reality of the automated harvesting of content on media sites in Quebec, in collaboration with the Quebec research chair on French-language artificial intelligence and digital technologies. The findings will be released in September.

We are taking a proactive approach to protecting right holders' content, but we should not be alone in speaking out against these violations. We want to make it clear that Copibec has nothing against artificial intelligence. We also want to point out that all creators have no issue with artificial intelligence, because they have always incorporated new tools into their work. That is not the issue here; the issue is to strike a fair balance in an evolving economic model.

We can't afford to wait for the outcome of the lawsuits against web giants in Quebec, Canada and other parts of the world, because artificial intelligence is advancing more quickly than the law can keep up. Creative industries are the pillar of Canada's culture and identity and a key economic driver. The sector cannot self-regulate. The government must enact legislation to protect our cultural sovereignty.

Thank you for your attention.

[The Vice-Chair, Conservative, Raquel Dancho]

Thank you very much, Mr. Laforce.

I'll turn the floor over to Mr. Buteau.

You have five minutes, please.

[Samuel Buteau, Consulting Program Officer, ControlAI]

Thank you, Madam Chair and members of the committee, for inviting me to testify today.

My name is Samuel Buteau. I have over a decade of technical experience in AI and now work at ControlAI, a non-profit organization that informs lawmakers of the risks posed by superintelligent AI and what the solutions are.

In March of this year, we submitted evidence to this committee outlining how the development of superintelligent AI represents a clear and grave national security threat to Canada. Since then, that threat has become harder to dismiss.

In April, Anthropic announced their new, most-advanced model, Mythos. During testing, Anthropic found that Mythos was able to gain unauthorized access to every major operating system and web browser, autonomously and outside of human control. Consequently, Anthropic took the decision not to release Mythos to the public, because, in their own words, “The fallout—for economies, public safety, and national security—could be severe.”

Nobody programmed Mythos to be the world's most capable hacker, because nobody programmed Mythos. AI systems are grown, not built, and nobody understands how they work. Anthropic's own CEO estimated that perhaps “we...understand 3% of how they work.” Importantly, that 3% figure is about what we understand today. Whilst Mythos is not being released publicly, it is still being used internally to develop the next generation of AI, and as AI development becomes automated, the ability for humans to understand AI systems and their capabilities will diminish even further whilst the speed of development rapidly accelerates.

We are currently on the path to developing systems that we do not understand and that are vastly more competent than humans in every domain. Experts say that superintelligent AI may arrive in two to five years. We will not be able to control these systems.

Mythos already threatens our national security, but we can manage it. It is clear that whatever comes in the future will not be manageable.

What can we do about this?

Firstly, the Canadian government should publicly recognize superintelligent AI as the national and global security threat that it is.

Secondly, just as it does for other national security threats, the government should begin closely monitoring the threat of superintelligence being developed and develop detailed scenario planning and doctrines.

Finally, but most importantly, Canada should take global leadership by forming a global coalition focused on preventing the development of superintelligent AI anywhere in the world. This coalition would form the basis for a mutual trust-but-verify regime where countries verify each other's supercomputers to ensure superintelligent AI is not being developed, whilst simultaneously monitoring the global flow of powerful chips.

Superintelligent AI is an equal threat to national and global security wherever it is developed. If China or any other country seeks to develop superintelligent AI, we must stop them at all costs. This trust-but-verify regime is an achievable and sustainable way to achieve this.

Thank you.

[The Vice-Chair, Conservative, Raquel Dancho]

Thank you.

We will go to our last witness, Ms. Luccioni, for five minutes, please.

[Sasha Luccioni, Co-founder and Chief Scientific Officer, Sustainable AI Group]

Thank you very much.

Honourable members of the committee, thank you for convening this crucial study.

As an AI researcher based in Montreal, my career has been dedicated to measuring the environmental impacts of artificial intelligence, first in academia, working at the Mila institute with Dr. Yoshua Bengio, but now in industry, where I recently co-founded the Sustainable AI Group, a research and advisory company dedicated to helping organizations measure and reduce the environmental impacts of the AI that they use.

In the last decade, my research has increasingly exposed a critical, often-ignored reality. AI is a very physical technology. It does not exist in an ethereal cloud. It relies on massive physical infrastructure. It consumes large amounts of energy and water, and it leaves behind a significant carbon footprint.

As Canada reshapes its digital policy agenda following the legislative reset of Bill C-27, we must understand that digital sovereignty and environmental sustainability are two sides of the same coin. Canada is uniquely positioned to lead the global shift towards responsible, sustainable AI. We should not attempt to outspend foreign monopolies on bloated general-purpose models and gigawatt data centres. Instead, our competitive advantage lies in building transparent, green and clean AI.

The integration of AI into sectors like construction and manufacturing does offer immense productivity gains, but the current industry trend of relying on massive, generic, cloud-hosted models is both ecologically and operationally unsustainable. The large language models that many of us use today are trained at great cost to be general purpose, which definitely makes sense from the point of view of the tech companies that want to respond to any kind of query thrown at them, be it coming up with a chocolate chip cookie recipe or an itinerary for a family vacation in Italy. These companies can and do spend hundreds of millions of dollars on compute in order to develop these models, often using our own data to improve them and then selling it back to us.

Most of what businesses in Canada need and want to use AI for isn't general purpose at all. It is specific tasks that require robust, trustworthy technology without all the bells and whistles. In fact, querying generic, multi-billion-parameter generative AI models to optimize a manufacturing assembly line or analyze a construction blueprint is like taking a helicopter to do your groceries. It is an absurd waste of energy. My research has found that generating a high-quality image with AI can use as much energy as half of a cellphone charge and that generating videos requires thousands of times more. The gap between the most and least efficient models is growing with the rise of reasoning models and agentic AI.

However, Canada has an opportunity to incentivize on-device AI and small, task-specific models that run locally within Canadian facilities. This creates a direct, powerful synergy between operational security and environmental sustainability.

Also, we can and should mandate transparency to audit the environmental and operational costs of training and deploying AI models. When we know where models are running, along with details regarding hardware and energy, we can accurately measure sustainability. The difference between training a multi-billion-parameter model on a low-carbon grid, like those in Quebec and Ontario, versus a grid powered by behind-the-meter natural gas can be orders of magnitude fewer emissions. Therefore, future federal AI legislation should mandate that companies deploying AI models disclose their full life-cycle environmental footprint, including the energy grid mix and water usage of the hosting data centres.

Canada can foster a booming local ecosystem by supporting companies that use open-source tools like CodeCarbon, which I helped develop, to publicly document their models' energy efficiency. This establishes a green AI stamp of approval, appealing to a global market increasingly desperate for ESG-compliant technology, differentiating Canada's AI offerings and giving our companies a competitive advantage. The federal government can lead the way by requiring AI developers to provide this information when applying for tender offers for government contracts, setting a baseline for the entire field.

Also, just as we rely on Energy Star ratings for appliances, Canada can implement energy certifications for commercial AI models, disincentivizing the use of loaded models for simple tasks. This approach is practical and highly achievable. It has already been proposed by the AI energy score project, which I have co-led for the last two years. In our work, we've tested hundreds of open-source AI models across dozens of tasks, finding efficiency differences in the tens of thousands between models of different sizes and architectures.

Finally, we must align the government's current industry plan with these ecological limits. The federal government's allocation towards the sovereign compute infrastructure program, SCIP, is a vital step toward reclaiming our data sovereignty, but compute power cannot be decoupled from environmental boundaries. The government must ensure that any public supercomputing infrastructure built under the SCIP is powered, at least in the majority, by renewable energy and utilizes cooling systems that sustainably use a local power supply and that are developed in consultation with local residents, including indigenous communities.

Instead of copying the hyperscale, football field-sized data centres that are becoming the norm in the United States, Canada has the opportunity to fund more creative approaches to compute. For instance, we can build smaller data centres that are better integrated with existing infrastructure, allowing us to reduce resource consumption while reusing the generated heat for offices, residences and university campuses.

Furthermore, SCIP resources should explicitly prioritize Canadian researchers and open-source initiatives developing climate tech solutions and sustainable industrial applications. Each project that plans to use SCIP compute should be required to measure and report their energy and emissions, improving the transparency of the field as a whole.

True digital sovereignty is completely impossible without environmental sustainability. Canada must reject the current trajectory of AI, which is unsustainable from all perspectives, and define a better trajectory for Canadian AI.

By legally mandating—

[The Vice-Chair, Conservative, Raquel Dancho]

Thank you very much.

Colleagues, we are running over time, of course, because of our delayed start. I'm going to conclude the meeting after the second round of the Bloc Québécois. We'll go around, and that will give us just over 30 minutes of questioning time for our excellent witnesses today.

We'll go to Ms. Borrelli for six minutes, please.

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

Thank you, Madam Chair.

Thank you to all of our witnesses today.

My first question is for Madame Desrochers.

AI systems have begun deciding what content is recommended, summarized, translated or generated for users. How should Canadian policy respond if AI systems become gatekeepers for what Canadians see and hear?

[Marie-Julie Desrochers]

I want to make sure I understand your question.

Is it about discoverability, if we speak about gatekeeping and the way the content is recommended? Is that the sense of your question?

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

Yes.

[Marie-Julie Desrochers]

When it comes to discoverability, legislation to modernize the Broadcasting Act was passed in 2023. The government of Canada took action to support the discoverability of content.

Indeed, artificial intelligence is a tool that can be used to improve the discoverability of content. When it is used for that purpose—to promote more content from Quebec, with proper metadata and without violating copyright or using protected works without the authors' consent and failing to compensate creators—it can be a positive form of use.

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

Is there a concern that generative AI could produce enormous volumes of low-cost music, scripts, images and text that will crowd out human-created Canadian works? How can we ensure that Canadian creators are not pushed out of their industries?

[Marie-Julie Desrochers]

Thank you very much for the excellent question.

There is a lot of concern on that front. It's actually happening already. I spoke with one of our members in the music industry, and he told me that over 75,000 artificial intelligence-generated tracks are uploaded on Deezer every day. That represents 44% of all new uploaded music daily. After they shared this data, the figure rose from 60,000 to 75,000 in just one month, and it keeps going up.

We are seeing this with music and with books. For example, you can find fake books on sale on Amazon that appear alongside recently published genuine works. We are also seeing this in the audiovisual sector. This is a genuine concern.

We have seen that some platforms, such as Deezer, are trying to clean up and are taking down and demonetizing this type of content. However, we are calling for mandatory transparency for artificial intelligence-generated content published on platforms to ensure consumers know what they are consuming and have the option to choose human-created works.

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

Thank you.

My next question is for Mr. Buteau.

Mr. Buteau, ControlAI has called for prohibiting the development of superintelligent AI and other dangerous AI, such as automated AI research or hacking capabilities. How should lawmakers define those prohibited capabilities in a way that's enforceable but not vague?

[Samuel Buteau]

Absolutely. It's a great question.

As AI development continues, AI is becoming more autonomous—able to independently take action and handle open work without human involvement. These are sometimes called AI agents. As I think everybody can appreciate, if we build systems that are much more competent than any group of humans in every domain, including hacking, persuasion, building new technologies, etc., and we don't control them, that's a bad thing.

How do you define and how do you create laws? First, obviously, human judgment would be involved. Just as with most laws, if a representative panel from signatory countries believes a system quite likely would be able to autonomously compromise national security, escape human oversight and upend international stability, we would call this superintelligent AI. If something could be trivially, covertly or quickly converted into superintelligent AI, that would count as well.

Third, there are two ways of developing superintelligent AI—on purpose and recklessly. I think here we should use the principle that we don't let people try to build a bomb, even if they don't know what they're doing. We also don't let people plot to overthrow the Canadian or U.S. government. That would be intent, but in the name of short-term competitiveness, one might still recklessly develop superintelligent AI by, for example, automating AI R and D.

The real thing that matters is this: What is the point of intervention? That's what you must define. We call those precursors. Two examples are large AI data centres and AI that can significantly automate AI research and development. There are others. We need to be flexible and be able to add precursors as they get discovered and as the landscape changes and so on. I think precursors are allowed, but they must be monitored within the regime and denied outside the regime.

[The Vice-Chair, Conservative, Raquel Dancho]

Thank you very much.

We'll go now to Mr. Bardeesy for six minutes.

[Karim Bardeesy, Liberal, Taiaiako'n—Parkdale—High Park, ON]

Thank you, Madam Chair.

I have a few questions for you, Ms. Desrochers and for you too, Ms. Luccioni.

Ms. Desrochers, the Coalition for the Diversity of Cultural Expressions has taken a strong stand on cultural exceptions in our free trade agreements and the government has said that we are strongly committed to that.

In your opinion, how important is this commitment on cultural exception when it comes to artificial intelligence and some of the challenges we are discussing today?

[Marie-Julie Desrochers]

Thank you very much for that question.

Indeed, we are paying very close attention to everything that is going on now during the review of the Canada—United States—Mexico agreement, or CUSMA. We need to protect our cultural exception because without it, we will lose some of our freedom to implement cultural policies of our own choosing domestically.

However, besides the exception, we need to safeguard the laws we have in Canada. For instance, the Broadcasting Act, which was modernized in 2023, must be non-negotiable. The same goes with the act respecting the discoverability of French-language cultural content in the digital environment, which was enacted in Quebec.

We need to be very vigilant when it comes to artificial intelligence and anything that will involve digital trade, which can be used to circumvent our exception because it may not directly touch on what is already covered within the scope of the exception and its definition. These issues should be looked at closely.

As it is, we have seen some companies in the United States publicly say that the CUSMA review provides an opportunity to secure exceptions for text and data mining, for example. We need to be very vigilant on that front to ensure that this does not happen.