Madam Speaker, as I was saying, there are many different uses. Did people consent to those uses? Did they consent to being in an ad when they were walking down the street? Did they consent to having their image captured as they were getting on a bus in Trois‑Rivières? Did they consent to their comings and goings being tracked? Did passengers arriving at Trudeau airport consent to being identified for their passport using facial recognition?
How is that data being used? Did people consent to that? Is that data protected? After the data leak at Desjardins, we have to wonder whether facial recognition data at risk too.
This is certainly something we need to work on. Unfortunately, the lack of regulations gives the bad guys a definite advantage. Those looking to sidestep regulations sometimes succeed, but it is complicated. However, if there are no regulations, then it is the wild west, really.
Let me give a few examples of the benefits that facial recognition technology has to offer. Daniel Therrien, the former privacy commissioner of Canada, said that facial recognition can be used to solve serious crimes, such as missing children cases. It can also be used for other compelling state purposes, such as in the border context to ensure that people of concern can be identified at the border while not impeding the flow of travellers to the country. Obviously, that is desirable. These kinds of uses are intended to protect us. I think we would all support them.
However, there are drawbacks, and they often concern mass surveillance. One thing is immediately obvious. Mass surveillance is definitely being done without a warrant. People are being surveilled at baseball games, on the bus and in the subway. Although the goal may be to identify a perpetrator, everyone is surveilled in the process. That is problematic.
As for the disadvantages, Patricia Kosseim, Ontario's information and privacy commissioner, told us that, with regard to facial recognition, the biggest concern of commissioners across Canada is mass surveillance, whether done by a third-party private sector company on behalf of the police or by the police service itself.
Assistant or deputy commissioners of the RCMP candidly told us that they were using facial recognition without a warrant and without letting the public know. Obviously, we can expect the RCMP to use facial recognition for legitimate, worthy reasons. However, the privacy commissioner also found that there had been serious and systemic failings to ensure compliance with the act before collecting personal information and before collecting information in general, for that matter.
I was talking about shopping malls a little earlier. I mentioned the buses in Trois‑Rivières because facial recognition is used on them too. I think we need to be careful, because on top of the inherent bias against Asian people and people of colour, for example, criminal bias exists as well. Poor-quality cameras can produce images that lead to a person being incorrectly identified. In short, facial recognition is not foolproof.
Now, our faces can be used for other purposes as well, including disinformation. We have started seeing videos on social media of Donald Trump and Kamala Harris kissing and partying together. They are public figures, but the same thing could happen to us. We could show up in a photo or video with someone who was never actually with us, doing something we never did.
We have to be careful. Disinformation is a serious problem today, one whose impact we often underestimate. All sorts of foreign actors can put information out there for all to see, thinking that they can convince people. Last week, I believe it was Tuesday, Communications Security Establishment Canada intercepted 6.6 billion attempts at disinformation in Canada. That is just another day at the office for the CSE.
The fact is that all of this information contributes to how we think. It may lead us to do things that we may not have done otherwise. That is a problem. Facial recognition is one many tools of disinformation.
There is another rather remarkable thing that is concerning. When it comes to the environment, we often hear talk of social licence. We need to be careful because social licence is a form of renunciation, for example, we would prefer A to B. Social licence does not necessarily equate to enthusiasm. However, there has never even been a debate about social licence or future social licence for facial recognition. It is assumed that, if we are in a public place, our face is part of the mosaic and that, if we did not want to be there, then we could just do nothing. In my previous career, people often told me that they had not done anything wrong and so it was no big deal if their image was being captured. I often answered those people by saying that, if they knew what could be done with those images, they might be more concerned. There are always malicious actors around, whether local or international.
The topic is not being discussed. We discussed it once with our colleagues on the Standing Committee on Access to Information, Privacy and Ethics. We discussed it with my hon. colleague from Barrie—Innisfil. However, these discussions have not necessarily filtered through to society as yet. Some groups have apprehensions, but no discussion is happening. Concerns are being raised, but that is not enough. Ultimately, we concluded that we should probably make a few recommendations. Given the total absence of any regulations, we had to at least come up with a few proposals that would make the use of facial recognition more transparent. I am going to quote a few of the 19 recommendations, including the first and foremost among them, which reads as follows:
That the Government of Canada amend...the Privacy Act to require a government institution to ensure that the practices [when using facial recognition]...are lawful.
We are talking about the Canada Border Services Agency, the passport office, a whole bunch of places like that. We figured that the Privacy Act had not been revised since before the Internet arrived on the scene, and that a little update would be in order. I am throwing this idea out to my colleagues on the other side of the House. All kinds of committee reports have been presented and, in its responses, the government often says that we have some good ideas. However, an intention without action is just an intention. It is worthless, even if it is a good one.
We also thought there should be clear sanctions for privacy violations committed by the police. After all, law enforcement agencies are among the biggest users of facial recognition. I am not blaming them; there are legitimate reasons for using it. However, when they do violate privacy, whether voluntarily or not, there should be clear sanctions. When an action has no consequences, people continue doing it because there is no cost, financial or otherwise. It becomes a habit.
The following is another one of our recommendations:
That the Government of Canada amend the Privacy Act to require that prior to the adoption, creation, or use of facial recognition technology, [the government] seek the advice...of the Privacy Commissioner...
The Privacy Commissioner needs to be consulted before a facial recognition tool is developed. This recommendation was made in 2022. In 2024, while Bill C‑27 is being studied in committee, people are still questioning whether Canadians need to be protected. It is right there, in black and white, in the report. We have to protect citizens because this data is not always used for legitimate reasons, and even if it is used legitimately, it is often used without a warrant. We have to be careful. I think this is a serious warning. To illustrate how important this is, the fact is that two years later, we are still talking about it. There have been no conclusions and, in fact, the situation has sometimes been trivialized.
I want to talk about another interesting recommendation. It is not often discussed. It is the right to be forgotten. Someone might want to be removed from the network. The European Union adopted a similar recommendation. The right to be forgotten is the possibility of contacting an agency that coordinates everything in order to allow an individual to not be automatically identified on social media or to be forgotten if they want to disappear. This may seem odd in a time of influencers who take selfies every four seconds, but a person may not want to be on the web for very legitimate reasons. We want the government to require “service providers, social media platforms and other online entities operating in Canada to delete all users’ personal information after a set period following users’ termination of use”. This could include responses to polls, text messages a person sends, or photos in which they are identified. We would like to make it possible for this to be deleted.
I will not quote the other 16 recommendations. By the very nature of the recommendations that were made following a lengthy consultation with the Privacy Commissioner, provincial commissioners and stakeholders who promote facial recognition, as well as those who criticize its use, there was unanimous agreement that something had to be done. As we know, nature abhors a vacuum, and where there is nothing, the nothing gets filled with something. It is frustrating.
Just before I close, I would like to quote a witness, Carole Piovesan, from INQ Law. She said that we need to be careful, that we need to increase transparency, but that, if we are going to do it, we need to do it “with a scalpel, not an axe”. The idea is to be aware of this relatively new technology, which, after two years, is no longer all that new. We can benefit from it, as I have just outlined, and we can guard against the harms, particularly the ones I mentioned.
