Canadian Heritage Committee on Oct. 29th, 2025

[The Chair, Liberal, Lisa Hepfner]

I call this meeting to order.

Welcome to meeting number 10 of the Standing Committee on Canadian Heritage.

Before we begin, I would like to ask all in-person participants to read the guidelines written on the updated cards on your table. These are measures in place to help prevent audio and feedback incidents and protect the health and safety of all participants, including the interpreters. You will also notice that there's a QR code on the card. It links to a short awareness video, if you need it.

Pursuant to the routine motion adopted by the committee, I can confirm that all witnesses have completed the required connection tests in advance of the meeting. That's for our two witnesses who are joining us online.

Please wait until I recognize you by name before you speak. All comments should be addressed through the chair.

Pursuant to Standing Order 108(2) and the motion adopted by the committee on Thursday, September 22, 2025, the committee is meeting to study the effects of the technological advances in AI on the creative industries.

We have with us today, as an individual, Michael Geist, Canada research chair in Internet and e-commerce law, Faculty of Law, University of Ottawa.

We also have Véronique Guèvremont, full professor, holder of the UNESCO chair on the diversity of cultural expressions.

We have Chip Sutherland, lawyer, joining us online.

We have Nikita Roy, from Newsroom Robots Lab, with us here in person.

We have Vicky Mochama, communications director from PressForward.

Welcome to all of you.

Each of you will have five minutes to present opening statements, starting with Michael Geist.

You have the floor.

[Michael Geist, Canada Research Chair in Internet and E-Commerce Law, Faculty of Law, University of Ottawa, As an Individual]

Thank you, Chair.

Good afternoon, everyone.

My name's Michael Geist. I'm a law professor at the University of Ottawa, where I hold the Canada research chair in Internet and e-commerce law. I appear in a personal capacity representing only my own views.

Thanks for the invitation to appear on this important study on AI in creative industries.

As some of you may know, I've appeared many times before this committee on questions involving technology and culture, including studies on copyright, freedom of expression and Internet regulation. In each instance, much of the discussion amounted to risk analysis, the perceived risk arising out of new technologies, whether digital copyright, online platforms, streamers or digital advertising, and concerns about risks associated with some of the proposed legislative responses, such as anti-circumvention rules, regulating user content or blocking news links. I think too often the debate frames new technology as a threat, emphasizes cross-industry subsidies and misses the opportunities that new technology presents. We therefore need risk analysis that rejects entrenching the status quo and instead assesses the risks of both the technology and the policy responses.

The debate over AI faces a similar challenge. I think that helps explain why the government has shifted from AIDA—the former Bill C-27—to now warning against overindexing on AI regulation and why groups that typically call for copyright reform find themselves arguing against it before this committee at the moment. These highlight the challenges of identifying AI risk and the fear that some regulatory responses could themselves create new risks that outweigh the problems they're trying to solve.

What are the risks I think this committee needs to think about with respect to AI and the creative sector? Three issues that often arise in this area are freedom to create, appropriate protections and Canadian content presence or discoverability. I think each of these presents its own challenges.

First, with respect to freedom to create, AI is already an integral part of the creative process, used to assist with everything from writing to film to music. Given its importance, AI has real benefits and restrictions on AI use are not only unrealistic but may be harmful. The risk comes from misinformation or public confusion that can come from “AI slop” in a video context or poorly crafted AI-generated news. This content should be properly identified, which would enhance the value of original human creativity. There is a need to work with the relevant sectors—news, video, music and AI services—to develop appropriate transparency measures to more easily distinguish between human-generated and AI-generated content.

Second, copyright invariably arises when discussing appropriate protections. Yet in the context of AI, the application of copyright isn't always clear cut. The outputs of AI systems rarely rise to the level of actual infringement given the expression may be similar or inspired by a source, but is not a direct copy of the original. The inputs—such as inclusion in large language models—are currently the subject of numerous lawsuits, but few have to date resulted in liability since those cases suggest large language models, LLM, inclusion and the resulting data analysis often qualifies as fair use or fair dealing.

What are the risks here? To paraphrase Minister Solomon, overindexing on AI regulation in a copyright context risks creating barriers that would render us uncompetitive as a market, undermining both innovation and creators. If Canada makes it more difficult or costly to develop large language models, AI development will shift outside of the country. It's therefore essential to ensure that our copyright frameworks are globally competitive. That's why we need copyright laws that continue to strike the balance through effective fair dealing rules and, given the use of text and data mining exceptions elsewhere, including the EU, the appropriate exceptions that position Canada as receptive to AI opportunities.

Third, we want to ensure AI services feature relevant Canadian results, but conventional Canadian content presence or discoverability policies such as minimum content requirements or promotional presence efforts simply don't map onto AI. Indeed, these kinds of policies could backfire, leading to the exclusion of Canadian content in large language models, which would in turn result in reduced presence in AI outputs. Essentially, it would be a replay of what we've seen with news on some social media platforms, where there are fewer conventional news sources and more presence of substitutable alternatives. In other words, the answer to Canadian AI cultural relevance is more Canada in the training data. That doesn't come from more regulation, legal barriers or higher costs, rather, it requires transparency on datasets, reducing costly barriers to access and the development of public AI systems that encourage the use and availability of Canadian content.

I look forward to your questions.

[The Chair, Liberal, Lisa Hepfner]

Thank you, Mr. Geist.

I now give the floor to Véronique Guèvremont for five minutes.

[Véronique Guèvremont, Full professor, Holder of the UNESCO Chair on the diversity of cultural expressions, As an Individual]

Thank you, Madam Chair.

Members of the committee, thank you for this invitation.

My name is Véronique Guèvremont, and I'm a full professor in the faculty of law at Laval University and holder of the UNESCO chair on the diversity of cultural expressions.

While I fully recognize the potential of artificial intelligence to support creativity and creative industries, my remarks will focus on some of its negative effects on the diversity of cultural expressions and on the cultural rights of individuals and groups. I will also go over Canada's international commitments that should encourage it to take action to limit those effects.

First, while generative artificial intelligence enables new forms of creativity and expands artistic opportunities, it also risks promoting homogenization and weakening cultural diversity.

Those risks stem primarily from the training databases for artificial intelligence systems. They're built from a corpus in which certain cultures are overrepresented, mainly anglophone and western cultures, which encourages the production of the same aesthetic or narrative forms.

Those models are designed to reproduce statistically average patterns based on their training data, which reinforces the most common styles. Non-conventional or marginalized forms of expression are often under-represented, which leads to representation biases in the creative content produced.

Other threats stem from the rapid increase in synthetic content in the digital environment. That overabundance could make human creations less accessible and less visible.

In the music sector, for example, such a scenario is mentioned in a European Parliament resolution passed in 2024, which emphasizes, “a growing number of [musical tracks] flooding streaming platforms on a daily basis, which risks aggravating existing imbalances as regards discoverability”. On some music platforms, artificial intelligence generates more than 10% of the tracks published each day.

Second, the impoverishment of the diversity of cultural expressions limits the right of individuals and groups to access their own culture. It also leads to an infringement of artistic freedom, particularly because of the increasing difficulties for creators to reach their audiences.

The impacts on the cultural rights of minorities and indigenous peoples are also evident, since the works of those groups are typically absent from training datasets, which limits their ability to use artificial intelligence to create works that are representative of their cultures.

Artificial intelligence systems also raise significant copyright protection concerns. As you know, while artificial intelligence systems can be used to generate new forms of cultural expression, they do so using works previously created by individuals and communities, generally without compensation for rights holders. This practice infringes on the right to benefit from the protection of the moral and material interests resulting from any scientific, literary or artistic production, a right enshrined in subsection 27(1) of the Universal Declaration of Human Rights and paragraph 15(1)(c) of the International Covenant on Economic, Social and Cultural Rights.

This obviously has an impact on artists' compensation. In addition, platforms are investing in AI-generated creation and promoting their own content to reduce copyright payments. A 2024 study published by France's International Confederation of Societies of Authors and Composers highlights, “Potential cannibalisation of creator’s revenue streams due to the substitution of human works by Gen AI outputs”. The study predicts that creators could lose nearly a quarter of their revenue by 2028.

Finally, I'd like to point out that ethical and legal frameworks relating to artificial intelligence generally stay silent on issues related to the diversity of cultural expressions and cultural rights. However, some of Canada's international commitments should encourage it to adopt an artificial intelligence governance framework that explicitly takes into account the points raised earlier, in order to limit their impact.

For example, UNESCO's convention on the diversity of cultural expressions, adopted in 2005, is complemented by operational directives adopted in 2015 that guide Canada's actions to protect and promote the diversity of cultural expressions in light of the rise of digital technologies in the cultural and creative industries.

The UNESCO Recommendation on the Ethics of Artificial Intelligence, adopted in 2021, recognizes that the rapid development of AI technologies challenges their ethical implementation and governance, as well as respect for and protection of cultural diversity. The recommendation calls on member states “to examine and address the cultural impact of AI systems”. Finally, policy area 7 of this recommendation, on culture, includes a number of relevant recommendations that will be useful to this committee's reflections.

I'm happy to expand on these topics during the discussion period. Thank you for listening.

[The Chair, Liberal, Lisa Hepfner]

Thank you very much.

Next I'll turn to Chip Sutherland.

You have the floor for five minutes, please.

[Chip Sutherland, Entertainment Lawyer and Artist Manager, As an Individual]

Thank you, Madam Chair.

Thank you for the opportunity to appear today before the committee.

By way of background, I am a lawyer based in Nova Scotia who has been practising entertainment law specifically in the music industry for over 30 years. I have represented artists at all levels both in Canada and internationally. In addition to my legal work, I also act in the capacity of artist manager for a select number of artists, currently including the singer Feist and Mustafa Ahmed. I co-authored the Canadian edition of the best-selling book on the music industry, All You Need to Know About the Music Business, with Donald Passman. I have also been the executive director of the Canadian Starmaker Fund for the past 21 years.

My purpose in being here today is to provide an insider's view of the life of the artist in the music business. As a lawyer and manager, I have the unique perspective of helping artists navigate their careers in all aspects of recording, songwriting and touring. Through my work with Starmaker, I have the advantage of seeing the development of all emerging artists in Canada. I know it's sometimes difficult in public proceedings to find the artist’s true best interest amongst a lot of the submissions from industry groups, but my career is built on finding and managing those best interests.

If you want to know how a musician really thinks and how they survive in an always-evolving commercial landscape, hopefully I can provide some of these insights.

You’ve heard from several thoughtful organizations at this point. I've seen some of the previous proceedings, so there is no need to repeat a lot of that. I think there is high-level agreement amongst the music industry participants that the introduction of generative AI is a significant threat to the creative industries and specifically the music business.

From the artist's perspective, I would suggest there are three areas where the Department of Heritage can provide leadership and support.

The first is strong artist funding supports. We currently have very strong supports for artists at all levels thanks to the Canada music fund and support from private broadcasters. In fact, we are the envy of many countries in this regard. The best defence for artists to combat and manage some of this new technology is to know they can count on the continuation of these strong supports at all levels. This is something the Department of Heritage can uniquely provide. The more we can support artists directly to sustain their ability to create music, the better. You can count on them to navigate the industry. I have seen musicians go from delivery of vinyl albums and recording on two-inch tape to producing CDs, home-studio recording on digital devices and, ultimately, streaming. I have seen them tied up in abusive recording deals all the way through to retaining ownership of their own masters and running their own labels. If we give them the support to create the music, they will figure out the best way to get it to fans.

The second is to strengthen some of the copyright provisions in this area. We need to reaffirm the copyright protection that exists now and reject any exemptions for data mining or scraping as being a fair use. Creative output is well protected right now by copyright, and all AI companies need to be reminded that any use of it without permission and compensation is stealing. I'm not saying you can't do it, but there needs to be a structure and framework for it. I believe you heard this from Patrick Rogers from Music Canada, who made this point.

The third thing to consider is the introduction and formalization of a personality right. In the U.S., there is a specific “right of personality” that is protected by their constitution. We don’t have an explicit right similar to that in our laws, although the courts have been inclined here to imply that a personality right exists in Canada. When someone steals the musical style of an artist, including their voice and maybe their image, it's a violation of not only the commercial rights and the copyrights but the very core of their character, likeness and personality. I think it would be helpful to review some of the laws relating to a right of personality and add further protections in that area.

Thank you.

[The Chair, Liberal, Lisa Hepfner]

Thank you, sir.

Next I give the floor to Nikita Roy from Newsroom Robots Lab.

You have the floor for five minutes, please.

[Nikita Roy, Chief Executive Officer, Newsroom Robots Lab]

Good afternoon, Madam Chair, and members of the committee.

My name is Nikita Roy. I am a data scientist, journalist and AI educator. I host the Newsroom Robots podcast and lead the Newsroom Robots Lab, which is incubated out of the Harvard Innovation Labs.

The work that my team and I do sits at the intersection of technology and journalism, helping media organizations worldwide harness AI to transform their work.

Artificial intelligence is becoming the infrastructure that mediates how people encounter information. It's redistributing the power to tell stories and to decide whose stories get heard. That transformation strikes at the core of Canada's cultural sovereignty and our ability to shape our own narratives in a world that is increasingly filtered by algorithms that we did not design and do not control.

Let me paint a picture of what's happening to our information ecosystem today. There are three shifts that are colliding all at once.

First, news is no longer something people just consume; they're starting to talk to it. Google, Microsoft and OpenAI are already having personalized, conversational news and informational experiences.

At Newsroom Robots, we built a voice-first AI expert for the news industry. It is trained on all of my own work and able to speak in over 30 different languages. It shows how AI can democratize access and also how quickly the ground is shifting. As news becomes something that literally anybody can talk to, we are entering a world where conversations guided by the AI algorithms—not simply articles—shape journalism. We must ask whose voices guide them and whose are left out.

The second shift is the collapse of search. For two decades, search was a major gateway to journalism. That gateway is now collapsing. We've moved from search, click, read and act to just ask, answer and act. AI cuts out the middle. There's no longer a home page, there's no click and there's no context. If Canadian journalism isn't built into those answers, it's invisible.

Across the world, increasingly searches now end without a single click. That is a silent collapse of the pathways that once led audiences to the news. Increasingly, the reader isn't even human; it's a bot that is training on journalism without consent and compensation.

The third shift is that AI is becoming the new home page. Platforms such as ChatGPT Pulse and Perplexity reassemble reporting from multiple newsrooms and present it within their own interface, stripping away our editorial voice. It's building a front page powered by our journalism without our bylines, without our context and without our curation. The Internet's front page is being rebuilt not by editors or publishers, but by algorithms of foreign AI platforms.

We are entering an era right now where the very infrastructure of knowledge is being rewritten by systems that we don't design and do not govern. If we fail to anticipate that cultural shift, our cultural sovereignty may not be decided in Parliament, but in the prompt and response loops of foreign AI platforms.

Together, these three forces—the collapse of search, the rise of AI interfaces and the shift from consumption to conversation—are redefining not just how we access information but who shapes the narratives that define us.

In this AI era, the greatest risk facing creators is invisibility. If our data, our languages and our voices aren't part of global models, we lose presence. We fade from the world's informational map. These aren't just traditional copyright questions; they are context rights questions—the right to be represented, visible and understood.

As UNESCO's “Artificial Intelligence and Culture” report warns, AI “is advancing faster than cultural governance, widening divides” and threatening cultural sovereignty.

Who gets to tell our nation's story, when machines become the interpreters of culture? If we want our creative voice to not just survive but to lead in this new era, we must act decisively on three fronts.

First is context rights. We must protect how our stories, languages and knowledge are used and understood within AI systems. Creators deserve transparency, credit and choice.

Second is capacity building. We must invest in creators' ability to work with AI and strengthen AI literacy across our creative and information sectors so that people can shape technology and not just be shaped by it.

Last year I launched and led the first-of-its-kind generative AI training for journalists at The City University of New York. The AI Journalism Labs, supported by Microsoft, since then has trained journalists around the world, including from Canada.

What I've seen is that when creators understand AI the right way, they stop fearing it and start shaping it. AI literacy is creative agency and that's what keeps culture alive in this new era.

Third is the Canadian data commons. We must build an ethically governed cultural data infrastructure, a kind of public library for the AI age that reflects our bilingual, indigenous and multicultural reality.

Our data is our cultural infrastructure, and our stories should not just become foreign imports in the digital age, because AI is not just changing how we tell stories; it's changing who gets to tell them, and that's what's at stake.

I'll give my time back to the chair. Thank you.

[The Chair, Liberal, Lisa Hepfner]

Thank you.

Finally, we have Vicky Mochama from PressForward. You have the floor for five minutes.

[Vicky Mochama, Communications Director, PressForward]

Thank you.

Good afternoon, Chair, and esteemed committee members.

I'm Vicky Mochama and I'm the communications director for PressForward.

Launched in 2020 with six founding members, PressForward has grown to represent 24 member media businesses across Canada. We advocate for, and organize with, media publishers, regardless of medium or business model, to ensure that independent media are heard on the issues affecting our work in the communities we live in and strive along with.

It should be noted that the news industry has been adopting AI tools for a fairly long time. For example, the Associated Press has been experimenting with AI since 2014. For our members, that means a variety of approaches and uses.

At Taproot Edmonton, in this year's municipal election, an AI tool helped the publication parse the issues most important to the city's residents, based on responses written to the publication by Edmontonians, and then shape a 30-question multiple-choice survey sent to all candidates.

For Future of Good, artificial intelligence tools help the publication in writing website code and with some copy-editing. Regardless, a human then edits and verifies the final product before publication.

At The Narwhal, a working group is consulting with readers and the newsroom alike on the publication's use of AI.

Each publication is responding to the AI question carefully, cautiously and with their communities in mind. Many publications are working across their entire businesses to craft responsible AI policies.

This will differ for each publication, its audience and its mission, but this can be said: There are uses of AI that are effective and ethical, and there are those that are destructive and unethical. I'll speak to the latter first.

As people who provide information, our journalism outlets rely on people in our communities coming to us for trusted, verified and bylined information. These points of connection are vital for a healthy public, particularly in the age of rampant disinformation and misinformation. These touchpoints are also where our publications generate revenue that covers the cost of journalism, which we do through advertising, reader donations or selling tickets to community events. AI summaries that reduce traffic to our sites by scraping human-verified information without compensation or attribution remove these points of connection.

As news publishers deeply embedded in our communities, our members highly value the trust and credibility that their communities place in them. We take responsibility for every single word, image and video that ends up on our platforms. We issue corrections, clarifications and apologies when necessary. In part, that is because news media businesses are legally liable for the veracity of their reporting. Practically speaking, this means there must always be a human in the loop to be held accountable.

Meanwhile, it's unclear what legal accountability generative AI, chatbots and the like can face when they get things wrong. Without the necessary guardrails and repercussions in place, the Canadian public are increasingly being asked to do their own fact-checking in a deeply fractious information ecosystem.

Moreover, we are in the business of news, while AI is often repeating what has already been established. The intellectual work of journalism cannot be replicated. AI tools may augment and assist, but cannot replace, the work of trusted journalists connecting with and reporting on what people are doing, thinking and saying in local communities.

In an era of misinformation and disinformation, toxic online discourses and digital foreign interference, trade wars and subsequent job losses, and indeed AI's as yet unresolved accuracy issues, it is even more essential to strengthen the media sector in Canada so that we can continue to serve our communities with factual information.

The Edelman Trust Barometer for 2025 notes that there is a “widening trust gap in Canadian society, with institutions facing mounting pressure to rebuild confidence and credibility.” PressForward's member publications play an important part in that rebuild by centring our communities and upholding the core values of transparency, accuracy and accountability.

In sum, AI is having a multitude of impacts on journalism in Canada, and there are of course some use cases where AI can be a useful tool in our media businesses, but generative AI specifically presents a number of threats and challenges to journalism by confidently distributing false or incomplete information; using news publications' content without payment or permission, or indeed attribution; and breaking the direct connection between our newsrooms and the audience.

In the last decade, the Canadian government has responded to this disruption and many others before it with supportive measures to shore up the production of quality journalism in Canada. We implore you to approach any AI strategy or regulation in a way that does not undermine those efforts.

Thank you, Chair.

[The Chair, Liberal, Lisa Hepfner]

Thank you.

Thank you to all our witnesses for being so on time today.

We'll start with questions starting with Mrs. Thomas from the Conservatives for six minutes.

You have the floor.

[Rachael Thomas, Conservative, Lethbridge, AB]

Thank you so much.

Ms. Roy, I was reading some of the things you've published, and I've listened to a number of podcasts that have been produced. One of the things you stated in an interview was the following. You said, “I believe that conversational AI is going to hold huge potential for news and it's going to be able to connect us to our audience even more than we've possibly been able to before.”

Obviously you're seeing a great deal of potential. You're seeing opportunity. I believe that you're fairly forward-thinking in terms of your willingness to consider AI and give it a role within the newsroom.

I hope that maybe you can expand a little bit on those opportunities that you see potentially playing in our future and how that can be done in a way that advances all Canadians.

[Nikita Roy]

The whole part I'm talking about, news becoming a conversation, is the big shift that we're going to see in how generative AI is going to revolutionize the way people access information. We're already seeing that with tech platforms, but if newsrooms are able to do it with trusted, verified information, what you essentially have is people engaging with news at a one-on-one level.

Right now, news is one size fits all. Imagine just for people across Canada itself. We're such a multicultural community, and we are able to cater to people across languages at the level of language they require. That's one way in which news is going to be able to help people be a more informed society.

The second part is that, when people are able to have conversations with the news, we are also able to understand a bit more what's missing. What aren't we covering? Right now when we publish a news story, we aren't able to engage with our audience other than through the comments section on our social media page, if you have comments available.

This transforms journalism from a one-way broadcast to a dialogue that we can have at an individual level. I think that will make for a more informed Canadian society. It's only when it's trusted information and within a trusted algorithm.

[Rachael Thomas, Conservative, Lethbridge, AB]

Right now, there are different news barometers that would say that this news source is more left of centre when it comes to the political spectrum, this one's more right, and this one's more centrist. You have this continuum of different news sources across the country.

My curiosity is this. Is there potential, then, that AI could bring a more balanced approach to news in giving the audience member the opportunity to access news from all sides or even present divergent views at the same time to the individual who's reading or listening to the source? Is that possible? Is there an opportunity there?

[Nikita Roy]

That's definitely a possibility. It's already happening.

If you look at a lot of the tech platforms when they are going and crawling newsrooms' data, they do in a way provide that information. If you are asking, “What does this newsroom versus that newsroom say? How do their words differ? How did their context differ?”, you're able to create all of that. That system exists.

The question now is who's in control of this algorithm and who's building that platform to do that.

There are also a lot of studies coming out where, if people believe in a particular type of conspiracy theory or something, a conversation with AI is able to help them understand the fallacies in their argument and help them understand what the truth is, in a way.

There are a lot of ways in which just conversation itself, that ability of having a back-and-forth conversation, helps people to understand information. The ability for AI right now to access a ton of information at the same time gives them an ability to see a spectrum of views.

[Rachael Thomas, Conservative, Lethbridge, AB]

If we as legislators want to ensure that the opportunity is protected, advancement is protected, innovation is protected and creativity is protected, we want to make sure that those things are possible. At the same time, we want to make sure that the dangers are minimized.

What would you suggest in terms of how that is achieved?

[Nikita Roy]

First of all, it is about getting the people who are doing that reporting compensated. If you go down to the first principles of what news is, news is a real-time structured knowledge network of verified information. Now, it can have different contexts and different framing, depending on the news site that you are going to. When an AI model kindly takes over that framing, that editorial judgment, that's the danger that I spoke about in my testimony. That's the danger; it goes away to foreign AI platforms, and it's not necessarily something that is controlled within Canada itself.

If you have, say, just a public broadcaster, the CBC, being that source of information where people are able to go, you're able to get it across different sides of understanding what the context is, what the framing is. That is where conversation helps people because, once again, you're able to tackle a particular story from different sides and opposing views and understand it that way.

I think what we need to do is really compensate people who are getting or producing that information, because it's possible right now, and it is happening already. I can go to ChatGPT or Perplexity and ask for the news across the political spectrum, but those people are not being compensated, and there's going to have to be a factor in the business model for those knowledge creators to continue producing that, because they will not be compensated.

[Rachael Thomas, Conservative, Lethbridge, AB]

Very quickly here, there are some who say AI replaces human creativity, and there are some who say that, actually, we can partner with AI for further creativity. Where would you fall on that spectrum?

[Nikita Roy]

Oh, completely on the pro-AI side.... AI does not replace human creativity; in fact, it amplifies it. Any person who did not even go to a journalism school or do an entire videography class today, if they have an imagination, is able to go and create it. I think that creativity is definitely amplified. I don't think it's going to replace it.

[The Chair, Liberal, Lisa Hepfner]

Thank you.

Mr. Al Soud, you have the floor now for six minutes.

[Fares Al Soud, Liberal, Mississauga Centre, ON]

Thank you, Madam Chair. It's great to see you, as always.

Thank you to our witnesses for being with us today. It's truly appreciated.

During the conversations we've had here, over the course of this study, we've started seeing themes evolve. The one I latch on to is the following: any potential cultural policy can't just protect the past, it has to govern the systems shaping what we see and what we don't see. I've often emphasized, at this committee, the importance of striking the right balance between preserving and empowering our creative sectors while also incentivizing innovation. This will resonate through my questions today, in general, as follows: How do we build a digital future where Canadian creators, institutions and cultural voices thrive, not in spite of new technology but through it?

With this in mind, Mr. Geist, you've previously said, “AI investment worldwide runs into the hundreds of billions of dollars, and the Canadian government contribution is never going to be more than a rounding error in total AI spending.”

You've long argued that our laws lag behind the realities of the Internet economy. This has come up on several occasions already. From your perspective, how does Canada create room for itself, given obvious barriers, while simultaneously protecting our cultural and creative sectors?

[Michael Geist]

You're right. I've been, at times, critical of some of the proposals that we've seen—including here at this committee. I should preface this by saying that I think your perspective, in saying how we facilitate innovation while at the same time adhere to some of our other objectives, is the right starting point.

The challenge we face is evident, even in the discussion we've had already so far. We've had a number of witnesses express real concern that the Canadian perspective—the wide range of views, languages, ethnicities and the like—somehow won't make its way in there, and so we won't be reflected in the outputs. At the same time, in the same conversation, we say we're really concerned about when you include our stuff, because that, then, raises concerns about whether or not we're being compensated.

One of the real things that we have to do, as part of even the broader government efforts around AI regulatory policy, is to say, “What do we really want to see happen here? Do we want to ensure that there is a Canadian perspective, there are Canadian virtues and values that are reflected in some of the outputs?" I think the danger is in erecting barriers and saying, “We don't want you to use our stuff.” It really comes at a significant cost, particularly for a committee that has been so focused on ensuring that Canada is well reflected in a wide range of media.

When I heard Ms. Roy talk about things like “transparency, credit and choice”, those are real, valuable pieces to start thinking about. I may have a different interpretation of it, but to me it's transparency, in terms of knowing where the sources are; credit, so that there's appropriate attribution since, sometimes, we do see people linking through; and choice, in a way that actually does empower sites, creators and others to ask, “Do you want to ensure your work is there, available and included in these large language models, or do you want to opt out?”

I think that, at the moment, we don't have a great system for it. I'll quickly give you an example. We've long had a system in “search” that basically says, “Search companies can index my content, but I can choose to opt out.” Many of the AI companies are, essentially, relying on that same signal, the robots.txt signal, and it seems to me that it's somewhat inappropriate here. It should be the case that a site should be able to say, “I want my content indexed for search purposes, but perhaps I don't want it indexed for large language model purposes.” We need systems that will allow people to make that kind of granular choice about who gets to index their content, in a sense.

[Fares Al Soud, Liberal, Mississauga Centre, ON]

That's very interesting. Thank you for that.

In your Globe and Mail piece, you noted that Canada faces a choice between U.S.-style deregulation and Europe's stricter rules. Do you believe there is a middle ground?

[Michael Geist]

Well, I think we are struggling to find one, to be candid about it. One reason for that—and this, of course, extends far beyond just the creative sector discussion and applies more broadly—is we know that capital and individuals are mobile. In the current environment, if we take a look at where the trends are headed—you talked about looking ahead—I think we do run real risks. If we establish regulatory frameworks that send the signal that we'd like to see AI here but only on our terms, we're going to find that many say, “Well, do you know what? We're going to shift elsewhere."

I think we've seen it quite profoundly, even with respect to the EU's approach, which, in a fairly short period of time, has gone from what some perceived to be the model for AI regulation to one in which many who believe that they want to be competitors in the AI space are increasingly wary of, because I think they have real fears that, ultimately, this will essentially extricate themselves from some of those opportunities.

[Fares Al Soud, Liberal, Mississauga Centre, ON]

Incredible.

I don't have that much time left.

Ms. Mochama, I'm quite curious: You've cited concerns about AI being a potential extinction-level risk, notably the tension between technological power and democratic communication.

From a communications and media viewpoint, how do we hold AI companies accountable to democratic principles when their tools shape public discourse?