Information & Ethics Committee on May 28th, 2019

[Ellen Weintraub]

Yes. It's my frame of reference.

[Ian Lucas]

Yes.

If I can talk to you, Mr. Cannataci, is the use of political advertising through broadcasting worldwide nowadays? Are there jurisdictions where broadcast advertisements are not accepted?

[Prof. Joseph A. Cannataci]

Thank you for the question, Mr. Lucas.

The answer is that national practices vary. Some countries go more towards the United States model. Others go towards the United Kingdom model. In truth, though, we are finding that in many countries where the law is more restrictive, in practice many individuals and political parties are using social media to get around the law in a way that was not properly envisaged in some of the actual legislation.

With the chair's permission, I'd like to take the opportunity, since I've been asked a question, to refer to something that I think is transversal across all the issues we have here. It goes back to the statement made by Ms. Weintraub regarding who is going to be the arbiter of truth. In a number of countries, that value is still very close to our hearts. It is a fundamental human right, which happens to reside in the same article 17 of the International Covenant on Civil and Political Rights, of which many of the countries around the table, if not all, are members.

In the same section that talks about privacy, we have the provision on reputation, and people [Technical difficulty—Editor] care a lot about their reputation. So in terms of the arbiter of truth, essentially, in many countries, including France—I believe there was some discussion in Canada too—people are looking at having an arbiter of truth. Call him the Internet commissioner or call him the Internet ombudsman, call him what you will, but in reality people want a remedy, and the remedy is that you want to go to somebody who is going to take things down if something is not true.

We have to remember—and this applies also to online harm, including radicalization—that a lot of the harm that is done on the Internet is done in the first 48 hours of publication, so timely takedown is the key practical remedy. Also, in many cases, while freedom of speech should be respected, privacy and reputation are best respected by timely takedown. Then, if the arbiter in the jurisdiction concerned deems that it was unfair to take something down, it can go back up. Otherwise, we need the remedy.

Thank you.

[Ian Lucas]

It occurs to me, going back to Mr. Erskine-Smith's point, that when I joined Facebook I didn't consent to agree to be targeted by political advertising from anywhere. I didn't know that was part of the deal. It's not why people join Facebook.

It seems to me that we've managed, give or take a few bad phases, to survive as a democracy in the U.K. without broadcast TV adverts, political adverts. It's a particular area of advertising that I'm seeking to restrict, and I'm supported because of the emphasis that was given this morning to the way the control of data is really removing choice from the individual in this process.

From an information regulation point of view, how clear do you think people are in that area? Do you think people understand that this is what's happening here?

[Ellen Weintraub]

First of all, let me say that I think there are many people in the United States who would love a system where they didn't get political advertisements. It's not something that people actually enjoy all that much.

I think you're raising a nuance there that I think is important, and that is, it's not.... Our Supreme Court would never allow a sort of a flat-out ban on political advertising, but—

[Ian Lucas]

Right. It's the position of the United States, but—

[Ellen Weintraub]

Right, but what you're talking about is the use of people's personal data to micro-target them, and that, it seems to me, does raise a very different issue. I'm not actually sure that it wouldn't pass constitutional scrutiny.

[Ian Lucas]

That's exactly what's happening in terms of paid-for advertising at the moment.

The other issue—

[The Chair, Conservative, Bob Zimmer]

Mr. Lucas, could you could make it real quick? I hate to cut you off. You've come a long way.

[Ian Lucas]

Very quickly, I'll pick up on an issue that Jens mentioned. He talked about trolls. Closed groups are also a massive problem, in that we do not have the information, and I think we need to concentrate more on that as an issue, too.

[The Chair, Conservative, Bob Zimmer]

Thank you, Mr. Lucas.

Next up, we have James from Ireland.

[James Lawless]

Thank you, Chair. I have another couple of questions and observations.

Just picking up on something that I think Nate was saying earlier in his other points, the question often is whether these platforms are legally publishers or dumb hosts, terminals that display the content that gets put in front of them. I think one argument to support the fact that they're publishers and therefore have greater legal responsibilities is that they have moderators and moderation policies, with people making live decisions about what should and shouldn't be shown. On top of that, of course, are the targeting algorithms. I think that's something that's of interest, just as an observation.

On my other point, before I get into questions, we were talking about nation-states and different hostile acts. One thing that's the topic of the moment, I suppose, is the most recent revelation in terms of the Chinese government and the Huawei ban, and the fact that Google, I think in the last few days, announced a ban on supporting Huawei handsets. But it strikes me that Google is tracking us through Google Maps and everything else as we walk about with our phones. I think I read that there are 72 million different data points in a typical year consumed just by walking about town with a phone in your hand or in your pocket. Maybe the difference is that somewhere Google has terms and conditions that we're supposed to take, and Huawei doesn't, but both are effectively doing the same thing, allegedly. That's just a thought.

On the legislative framework, again, as I mentioned earlier, I've been trying to draft some legislation and track some of this, and I came to the honest ads act. One of the issues we've come across, and one of the challenges, is balancing free speech with, I suppose, voter protection and protecting our democracies. I'm always loath to criminalize certain behaviours, but I'm wondering what the tipping point is.

I suppose that in the way I've drafted it initially what I've considered is that I think you can post whatever you whatever you want as long as you're transparent about who actually said it, who is behind it, who is running it or who is paying for it, particularly if it's a commercial, if it's a paid-for post. In terms of the bots and the fake accounts, and what I would call the industrial-scale fake accounts, where we have a bot farm or where we have multiple hundreds or thousands of users actually being manipulated by maybe a single user or single entity for their own ends, I think that probably strays into the criminal space.

That's one question for Ms. Weintraub.

I suppose another question, a related question, is something that we struggle with in Ireland and that I guess many jurisdictions might struggle with. Who is responsible for policing these areas? Is it an electoral commission? If so, does that electoral commission have its own powers of enforcement and powers of investigation? Do you have law enforcement resources available to you? Is it the plain and simple police force of the state? Is a data protection commissioner in the mix as well? We have different types of regulators, but it can be a bit of an alphabet soup, and it can be difficult to actually pin down who is in charge. Also, then, if we do have somebody in charge, it can be difficult; they don't always have the resources to follow through.

That's my first question. In terms of criminalization, is that a bridge too far? Where do you draw the line? Second, if there is criminalization and there's an investigation required, what kind of resourcing do you have or do you think is needed?

[Ellen Weintraub]

Again, my expertise is in the U.S. system, so I can most effectively tell you about that. We are a law enforcement agency. We have jurisdiction over money and politics, and we have civil enforcement authority. We have subpoena authority. We can do investigations. I think that some of our enforcement tools could be strengthened, but we also have the ability to refer to our justice department if we think there are criminal violations, which are basically knowing and wilful violations of the law.

Going back to something that you said earlier, in terms of our regulatory system, I don't think bots have first amendment rights. They're not people, so I don't have any problem with.... I don't understand why these smart tech companies can't detect the bots and get them off the platforms. I don't think that would raise first amendment concerns, because they're not people.

[James Lawless]

Actually, that's a great line. I'm going to use that again myself.

I think I still have time for my next question. There is another way around this that we've seen in Ireland and, I guess, around the world. We've heard it again today. Because of the avalanche of fake news and disinformation, there is a greater onus on supporting the—dare I say—traditional platforms, the news media, what we'd call independents, quality news media.

There is a difficulty in terms of who decides what's independent and what's quality, but one of the approaches that we've been looking at I think I heard it in the Canadian Parliament when we watched the question period a few hours ago. I heard similar debates. One solution we're toying with is the idea of giving state subsidies or state sponsorship to independent media, not to any particular news organization but maybe to a broadcasting committee or a fund that is available to indigenous current affairs coverage, independent coverage.

That could be online, or in the broadcast media, or in the print media. It's a way to promote and sustain the traditional fourth estate and the traditional checks and balances of democracy but in a way that I suppose has integrity and is supported, asks questions of us all, and acts as a foil to the fake news that's doing the rounds. However, it's a difficult one to get right, because who decides who's worthy of sponsorship and subsidy and who isn't? I guess if you can present as a bona fide, legitimate local platform, you should be entitled to it. That's an approach we're exploring, one that has worked elsewhere and has seemed to work in other jurisdictions.

[The Chair, Conservative, Bob Zimmer]

Thank you James.

Next we'll go to Mr. Graham.

Go ahead for five minutes.

[David Graham, Liberal, Laurentides—Labelle, QC]

Thank you.

Ms. Weintraub, to build on that, if bots don't have first amendment rights, why does money?

[Ellen Weintraub]

Well, money doesn't have first amendment rights. It's the people who are spending the money who have first amendment rights.

Let me be clear about this. I'm not a big fan of our Supreme Court jurisprudence. I mean, I would adopt the Canadian jurisprudence on this stuff if I could, but it's a little bit above my pay grade.

[David Graham, Liberal, Laurentides—Labelle, QC]

Fair enough. So you don't necessarily think the decision in Citizens United was a good one?

[Ellen Weintraub]

It would be fair to say that I am not a fan of the Citizens United decision.

[David Graham, Liberal, Laurentides—Labelle, QC]

Fair enough.

As you were saying, the elections commission's job is to oversee the financing of elections. If a company knowingly permits the use of its algorithm or platform to influence the outcome of an election, would you consider that to be a regulated non-monetary contribution?

[Ellen Weintraub]

I think it could be an in-kind contribution.

[David Graham, Liberal, Laurentides—Labelle, QC]

Interesting. My next question is related to that.

The CEO of Facebook has a majority of voting shares in the company. He basically has absolute powers in that company. From a legal or regulatory point of view, what prevents that company from deciding to support or act in any way they feel like in an election?

[Ellen Weintraub]

As a corporation, it can't give a donation directly to a candidate, and that would include an in-kind contribution.

A question was raised earlier—and forgive me as I can't recall if it was you who suggested it—about Mark Zuckerberg running for president and using all of the information that Facebook has accumulated to support his campaign. That would be a massive campaign finance violation because he doesn't own that information. Facebook, the corporation, owns it.

The wrinkle in that is that due to the decision of our Supreme Court, corporations can make contributions to super PACs, which supposedly act independently of the campaigns. If a super PAC were advancing the interests of a particular candidate, a corporation—including Facebook—could make an unlimited contribution to that super PAC to help them with their advocacy.

[David Graham, Liberal, Laurentides—Labelle, QC]

Understood.

I don't have much time, so I'll go to Mr. Therrien for a quick second.

To Mr. Lucas's point earlier, in the social media world, are we the client or are we the product?

[Daniel Therrien]

It is often said that when you do not pay with money, you are the product, and there is certainly a lot of truth to that phrase.