Information & Ethics Committee on May 7th, 2024

[René Villemure, Bloc, Trois-Rivières, QC]

Thank you very much.

[The Chair, Conservative, John Brassard]

Thank you, Mr. Villemure.

Thank you, Mr. Loewen.

Mr. Green, I believe that's you at the end of the table, way back there. You have six minutes. Go ahead.

[Matthew Green, NDP, Hamilton Centre, ON]

Thank you very much, Mr. Chair.

I'm going to spend the bulk of my time centring in on and trying to get the best recommendations we can from the witnesses. I'm going to be putting some questions to the witnesses.

I'll ask that you try to answer in a succinct way in order to allow for other witnesses to opine as well. If I ask the questions in a rapid-fire way, it's not me being curt; it's just me being judicious with my time.

I want to begin with you, Mr. Al-Rawi. You proposed a non-partisan fact-checking commission or body that might be able to separate out facts from opinions when it comes to misinformation and disinformation from MPs. Is that correct?

[Dr. Ahmed Al-Rawi]

That's correct.

[Matthew Green, NDP, Hamilton Centre, ON]

We heard in subsequent testimony that there is the ability—through both Mr. Loewen's media ecosystem, which collects data, and Mr. Frank's work around the dark crawler and the dark web....

Could you see this being taken up in an effective way by AI tools, or is this something that you would see simply as being professional people and subject matter experts in a human context, trying to keep up with the scale of all the misinformation and disinformation?

[Dr. Ahmed Al-Rawi]

Thank you very much for the question.

From my somewhat limited experience working with AI tools—even the new ones—I can be certain saying we haven't reached that stage yet. We still need humans to qualitatively assess pieces of information. There are cues that could be easily detected with AI tools—for example, if an image is created by AI technology—we call them deepfakes—or a video is being deepfaked. However, there must be some kind of qualitative assessment done by humans.

Thank you.

[Matthew Green, NDP, Hamilton Centre, ON]

To round it up with you, sir, do you have other recommendations for legislation?

If you were AI, I'd prompt you by saying, “Pretend you're an MP. Give us the best recommendations you can to help counter some of the gaps and some of the threats you've identified.”

[Dr. Ahmed Al-Rawi]

I think this is a collective effort. I don't want to say that parliamentarians should do all the work alone. Everyone should be involved. I know that some NGOs were invited to this committee. I think they should also be involved in this work.

We need to fact-check each other, actually, including myself.

[Matthew Green, NDP, Hamilton Centre, ON]

I'll re-ask the question. I'll re-prompt you.

We're here as members of Parliament. At the end of this study, we're going to be examining the testimony. The testimony will not consist only of just what's wrong—we've spent a lot of time talking about what's wrong, and we're probably still only scraping the surface—but what we have to get from the testimony are recommendations.

Understanding what our powers are, what our mandate is as a committee, what would you recommend to us that we adopt in our final report to help offset—certainly not to solve; I'm not talking about a silver bullet—some of the challenges that you've outlined?

Then I'll put that question to the two other witnesses.

[Dr. Ahmed Al-Rawi]

I'm not sure what to say here, but I believe the fact-checking initiative could be very useful because—

[Matthew Green, NDP, Hamilton Centre, ON]

Okay. I will take back my time and I will go on to Mr. Frank. I do appreciate somebody who says they don't have the answer. I have to say that all the time.

Mr. Frank, from your perspective, what is it you would recommend to us for consideration in a final report on this topic?

[Dr. Richard Frank]

I have lots to say, a lot more than the amount of time here, but there are a couple of main points.

This has to be done with the help of AI. What we're seeing right now is just a preview. This is going to get significantly worse as the disinformation is going to be AI-generated. The use of AI eventually will have to be done to detect this content, to de-escalate it and to intervene.

During our studies, we've always employed humans, domain experts, people of specific communities in which we wanted to detect disinformation. The approach we've used, I think successfully, was to get members of the community to point out examples of disinformation topics and then use that to start to train an AI model, which then can pick up on this and continue detecting new sources.

[Matthew Green, NDP, Hamilton Centre, ON]

With that model, if I'm to understand—and you can just say yes or no—is this the dark crawler model that you put into the dark web?

[Dr. Richard Frank]

It's within that context, yes.

[Matthew Green, NDP, Hamilton Centre, ON]

Given the vastness of that, could an application not be set upon us, as the 343 seats that will happen in the next election, including political parties? It seems like a much smaller digital ecosystem than the dark web.

[Dr. Richard Frank]

It has to be done within each community. If it is the members of Parliament, that would be one community. We would need experts who know what's going on, who would be able to start pointing to the initial disinformation.

[Matthew Green, NDP, Hamilton Centre, ON]

I have 30 seconds left.

Professor Loewen, could you pontificate on recommendations?

[Peter Loewen]

I would say, Mr. Green, we've decided in our country, through a lot of legal wrangling, to allow the limiting of speech during elections. We limit it to politicians largely and to parties, and we limit how much third parties can speak. That opens a legislative door for you to decide on what the arena will look like during elections.

The challenge is that generating speech through AI is virtually costless, and it can be done by people who are not people. The legislative framework that tries to limit which people can speak and limits how much they can speak through money is not fit for purpose for the world that Mr. Frank in particular is describing. You need to find a way legislatively to try to maintain the equilibrium that we have now in this new environment.

[Matthew Green, NDP, Hamilton Centre, ON]

Mr. Loewen, I will be coming back to you in my second round. I look forward to more insight on that.

[The Chair, Conservative, John Brassard]

Thank you, Mr. Green.

That concludes our our first six-minute round.

We're going to move to five-minute interventions.

I have Mr. Brock, followed by Mr. Bains, and then two and a half minutes for Mr. Villemure and two and a half minutes for Mr. Green.

Mr. Brock, you have five minutes.

Go ahead, please.

[Larry Brock, Conservative, Brantford—Brant, ON]

Thank you, Chair.

Thank you to the witnesses for their attendance today. This is interesting information.

Before I get into my questions, Mr. Frank, in particular, I noted you ran out of time during your opening remarks. I'm willing to give you as much time as necessary. Do you think you could wrap things up in under a minute?

[Dr. Richard Frank]

Thank you.

I was going to suggest a couple of positives for what we should be doing. Again, I do appreciate the time.

My first suggestion would be to establish trust in journalistic sources, or somehow help establish trust. I think we're losing it. A lot of people are getting their news on social media. We need to pull them back into trusted sources. Once we do that, I think people will be more robust in standing up against disinformation, but that needs proper funding. At the same time, the government can't be seen as the arbiter of truth, so this has to be positioned carefully.

I've developed courses on my own for educational campaigns. We just need to make people question more what they're seeing to make sure it is real.

The government pledged a lot of money for research. That's excellent. We need that. I think we need to bring AI into the discussion.

[Larry Brock, Conservative, Brantford—Brant, ON]

Thank you for that.

I'm going to stick with you, Mr. Frank, because my colleague Mr. Barrett sought remarks from Mr. Loewen with respect to his perceptions of Justice Hogue's interim report on foreign interference.

I think all three of you will agree that the misinformation and disinformation campaign, which has been happening for several years now, and at least throughout the 2019 and 2021 elections and currently in this country, has substantially impacted Canadians' confidence in our democratic institutions. That is something that we all, as parliamentarians, have to work on.

I will quote a couple of paragraphs from Justice Hogue's report and, Mr. Frank, I'd like to get your commentary on it.

In your opening remarks, you spoke specifically about the candidate Kenny Chiu and the impact of the disinformation and misinformation campaign, not only from Communist China but as well as the Liberal candidate who ultimately won the election and his participation in the process. I believe you also made reference to the fact that you are aware of Kenny Chiu's testimony at this committee. Is that correct?

[Dr. Richard Frank]

I know some of it, yes.

[Larry Brock, Conservative, Brantford—Brant, ON]

I'm going to quote a couple of paragraphs from an article. In the case of former Conservative MP Kenny Chiu, the commissioner identifies that there were “strong indicators” of Beijing's interference campaign and that there was a “reasonable possibility” that this resulted in the defeat of Mr. Chiu and the election of his Liberal opponent.

The article states:

It is also clear that as Canadians judge the actions of the Trudeau Government and its failure to prevent this interference, that we must consider Commissioner Hogue's conclusion that the interference that was allowed to occur undermined public confidence in our elections. She notes that the risk of foreign interference in our elections will increase as long as the Liberal Government fails to take “sufficient protective measures to guard against it.”

What are your comments, Mr. Frank?