Industry and Technology Committee on June 8th, 2026

[Jon Camfield]

Yes. We maintain a scam prevention hub with exactly that sort of content. It's not always localized to a community as targeted as you mentioned, but it is localized to countries. We also have a series of newsroom posts talking about specific targeted scams that we see. We work with external research firms to look at different communities and different trends. We had one recently on elder fraud and on current trends and tactics targeting elders around the world. We had one on romance scams and so on.

[Randy Hoback, Conservative, Prince Albert, SK]

You're talking about this hub. I didn't know it existed. I don't think anybody in this room knew it existed. What are you doing to proactively disclose that you have this hub?

Are you actually shaming these scammers? If you come across a scam that's in, say, another country, are you able to actually go after those people in that country? Can you say, “Hey, you're scamming in Canada, so we're going to take you out”, whether that be in Asia or wherever they're coming from?

[Jon Camfield]

I have two things. On the proactive side, one of the tools we use, which is actually one of my favourite tools, is a series of just-in-time warnings that we deliver to users, whether it's in messaging or with a new friend contact. We'll say, “Hey, this user isn't part of your social graph. They say they're coming from here, but we don't think that's where their location is actually based.” Any one of these signals is easy for a scammer to get around, break, and get through, but we can feed a collection of them to the user and say, “This looks suspicious. It could be legit, because it could be your friend travelling, but you should have your guard up.”

That's really the key to prevention. It's about not focusing on one specific trend, because the more you do that, the faster the scammers will move. It's to really focus on that—

[Randy Hoback, Conservative, Prince Albert, SK]

I agree with that. “Buyer beware” also has to be a factor in all this. It's just like the news on Facebook: “Watcher beware”.

You talked about 90% of your advertisers being verified. Where I have a problem with this—I agree that they're verified—is that the posting they're advertising with is garbage. It's an influencer who's getting paid by somebody else in the background. It can be the Lego guys from Iran or groups like that. All of a sudden, my face is being posted in that reel.

What are you doing to allow me to make sure that I can't be part of that reel and I don't end up endorsing a post that we know is actual BS?

[Jon Camfield]

We have multiple different ways to do impersonation protection, particularly for public figures. We can—

[Randy Hoback, Conservative, Prince Albert, SK]

It's not impersonation in this case. You're taking my name without my consent and saying I'm endorsing something. That's problematic to me, especially being a public figure.

My friend says, “I didn't endorse that”, but everybody in their media platform tells them, “No, you were on that platform endorsing it.” Is that good corporate behaviour?

[Jon Camfield]

I would say that our reaction to it and our tools to defeat and deal with it are very powerful and are of good behaviour. We work with brand owners and IP owners, including celebrities and public figures, to give them tools to report, identify and proactively detect—

[Randy Hoback, Conservative, Prince Albert, SK]

This is the average person. It's my neighbour down the street. I think he's an honest guy, and all of a sudden he's endorsing something that he's never even seen before. You've yanked his information and put it into that reel.

You talked about things changing. The other thing I'm seeing right now is fairly new. It's where the newscaster looks really real, but it's AI 100%. They're talking about a current event, and there's enough accuracy on that current event that you think it's true. It sways your opinion, but it's all BS. That's coming at people more and more. They can't determine what's real and what's not.

Is it fair to say that Facebook should be viewed as just an entertainment platform and not be viewed as a newsworthy platform?

[Rachel Curran]

I don't think it's news versus entertainment. It's that where that violates our policies and where it's misleading people and providing misinformation, we should be taking that down, or at least be applying a label to it so that people understand that it's generated by AI, it's not real and they ought to view it with some degree of skepticism. Those are all the tools we're deploying now.

Look, I agree that we can and should do more on this front, but where people are employing AI-generated deepfakes, for instance, or videos that are AI-generated, we require them to use labels and to label that as AI content. When it presents a significant risk of misleading the public, we will take down video material or AI-generated material, or we will put a label on it saying that it is not true and it should be viewed with significant skepticism.

[Randy Hoback, Conservative, Prince Albert, SK]

I have one question on that, then. I don't want to talk about the Online News Act. We've talked about that enough and we have our opinions on it. With your proactive activities in trying to take this off your platforms, what legislation or regulations—what else—are there that could allow us to empower groups to have more control of what's being put on your platform so that you can make sure that I'm not endorsing somebody I don't want to endorse or that you have the tools to remove these people more quickly?

I also want to know what you are going to do. You talked about your hub. Nobody knows about it. You need to do a better job of getting out there and identifying scams.

For the last thing, I'll use an example. On Facebook Marketplace, there's a set of patio furniture for sale in the next town. It looks beautiful. It's super cheap. It's too good to be true. When I look in the background of the picture, there are palm trees. I'm in Saskatchewan. Now, we're not as cold as Winterpeg, but how is AI going to interplace...?

You have two questions here. Get legislation that could actually open the door and make it easier to get after these folks. We need to punish the person doing it, somehow. We need to figure that out. Also, on the preventative side, how do we identify things quickly and say, “No, this is not real”?

[Jon Camfield]

In terms of legislation, I absolutely agree with my colleague on the protective measures for information sharing. Enabling us to do real-time signal sharing will get a lot of incredible benefits. As I mentioned, the result of any one seed we get back from banks is that we're able to take 600 times more assets down, which are not necessarily active assets. They could be completely dormant and not being used but will be used at some point in the future. That gets into some preventative measures as well.

In terms of direct enforcement, that's the value we see coming from government interaction. Sharing with law enforcement and having law enforcement share with their counterparts around the world have been incredibly powerful. That results in real arrests of real scammers, particularly when they cross borders seeking medical care or R and R. Otherwise, they are operating with impunity, from Cambodia, Myanmar and other places where there's no functional enforcement of their bad behaviour. Sharing that information gives a chance to actually make an impact.

[Randy Hoback, Conservative, Prince Albert, SK]

I have another little one.

[The Chair, Liberal, Ben Carr]

Ask it very quickly.

[Randy Hoback, Conservative, Prince Albert, SK]

You talked about law enforcement. Have you been proactively working with law enforcement to create a system in which they can plug into you quickly so that when they come across a scam—in this case, little Johnny being arrested for scamming—they can get that out into a region? You have the ability, through advertising, to say, “We're only going to advertise in this region.” Are you doing that with police forces? Are you doing that with local RCMP, local city police and county sheriffs?

[Rachel Curran]

Yes, we are—regularly.

I can speak to what's happening in Canada. We're working with the RCMP and with regional and municipal police forces on issues like scams. We liaise with them regularly about what we're seeing. What they're seeing, they report to us. We have a special portal for law enforcement through which they can report to us, and if there's an emergency request, they can send it to us very quickly. If it's not an emergency request, we can work together on how to take material down and address it over the long term.

[The Chair, Liberal, Ben Carr]

Thank you very much, Mr. Hoback.

Ms. Begum, I understand that you may be splitting some time with Madame O'Rourke. The floor is yours.

[Doly Begum, Liberal, Scarborough Southwest, ON]

Thank you very much, Chair.

Good afternoon—or good evening, almost—to all of you. Thank you so much for being here and for giving us the opportunity to ask you questions.

I'll start with Ms. Curran.

Do you think Meta has a solid enforcement mechanism for tracking down misinformation, disinformation, scams and frauds at this very moment?

[Rachel Curran]

These are all different issues—misinformation and disinformation, yes, and scams, yes—but there are always gaps between what we prohibit under our content policies and what we enforce against. We're constantly working to try to close that gap.

[Doly Begum, Liberal, Scarborough Southwest, ON]

In that case, from what MP Bardeesy was talking about, it seems that you were able to take down a Globe and Mail article very quickly. Meta does have an enforcement mechanism to take something down quickly, like an article that was harmless and in fact would have actually added to the quality of life of people and society.

I want to quote the Reuters article that was mentioned by Mr. Shipley and others here. The unreported document showed that Meta “failed to identify and stop an avalanche of ads that exposed Facebook, Instagram and WhatsApp’s billions of users to fraudulent e-commerce and investment schemes, illegal online casinos, and the sale of banned medical products.”

You're able to track down a Globe and Mail article that fast, yet you have an avalanche of ads that exposed billions of users. Do you not have mechanisms for that?

[Jon Camfield]

We do. The vast difference is the scale of the adversary. It's about how dedicated and well funded the scammer industrial complex targeting our systems is.

[Doly Begum, Liberal, Scarborough Southwest, ON]

Speaking of funding, it was projected that Meta has earned 10% of its annual revenue in the last years from scam ads.

[Jon Camfield]

The 10% number is a misrepresentation of the actual research.

[Doly Begum, Liberal, Scarborough Southwest, ON]

What was the percentage?

[Jon Camfield]

From the research at that time, it was in the 3% to 4% range, not 10%.

[Doly Begum, Liberal, Scarborough Southwest, ON]

Okay. I have between 5% and 10%. Even then, if we add up the ad revenue by taking, let's say, your 5%, that's about $6 billion in revenue that Meta made just from ads that were scams and fraudulent. People were getting scammed. Seniors were getting scammed.