Amazon's AI Face Tracking Laws?

AI Face Tracking Laws?!

Sebastien: What up what up? Welcome to The Digital Acid, the only podcast on the internet where you can go to find the best of future tech and street culture. I'm your host for today, Sebastien, and I'm joined by my homie Nathan go on and say what up!

Nathan: What’s good? What’s good?

Sebastien: We're going to be talking about some crazy mind-blowing topics today. Namely, that be Jeff Bezos says Amazon is writing its own facial recognition laws to pitch the lawmakers. So we're going to be talking about that crazy topic, and more.

Sebastien: But first, we have a quick message from our sponsor. Starfish Community is a co-working space and community hub for emerging technology pioneers to connect, collaborate and build. They host blockchain crypto, artificial intelligence, machine learning, IoT virtual reality and augmented reality, and other builders in the emergent webspace. They do some awesome events and if you'd like to find out more about Starfish Community you can visit their website at StarfishCommunity.com

Sebastien: So it seems like the richest man in the world wants to write laws for facial recognition to pitch the lawmakers. What's your hot take on that?

Nathan: NO. I'm gonna come in with the big NO.

Sebastien: With the hard NO?

Nathan: Yeah, I think somebody shouldn't be writing laws about this, but I don't know if it should be Amazon.

Sebastien: Yeah, I mean, I think it makes sense that they would want to be first that space. You know, being I guess, devil's advocate, there are very few companies that could probably speak to the nuance, and what might limit facial recognition technologies. Let’s say, you know, for example, like the way that a lot of the facial recognition happens is huge databases of faces that are public and compiled.

Sebastien: So let's say that they were to say “You can no longer source public data to create an algorithm” or something like that. Right? That may limit, you know, computer vision technology as a whole. So to be devil's advocate, it does make sense that technology companies would have a say as to what would be overstepping and over-regulation. But I'm in agreement with you. Like certainly I don't think the only proposal at the table should be Amazon. Yeah.

Nathan: Nah, come on dude! That's messed up. He can't be out here writing laws. I feel like a lot of that is going to be like in self-benefit, not really for the greater good. You know what I'm saying?

Sebastien: I think it's reasonable to think that companies would act in their own interest. So that makes sense. Now here in San Francisco, actually, the government was quick to ban facial recognition technology being used within the city for like, I believe law enforcement purposes.

Nathan: Mhmm. Yeah, I'm gonna stand by that.

Sebastien: Yeah, I'm in agreement.

Nathan: Right.

Sebastien: And so it can be helpful...

Nathan: But that's just an invasion of my privacy man. I don't want you to have like, computer vision at any given time that can just recognize who I am. You don't need that.

Sebastien: Well, here's the thing.

Nathan: That’s over-stepping.

Sebastien: So let me propose a few things. You probably have a phone that can recognize your face. And facial recognition can be applied using artificial intelligence through any cheap webcam, any cheap camera. So that means that the one that you carry around all the time has the ability to do that. And pretty much anywhere where you could see that being applied in CCTVs in the future. China has a huge facial recognition thing with what’s going on in Hong Kong right now. People are actually applying masks over their face that project different faces on top of their face to distort facial recognition systems that the government is using to prosecute people.

Nathan: I mean, how else like I would probably be wearing that to dude!

Sebastien: Right.

Nathan: Not that I have a reason to be prosecuted, but I just don't want that invasion of my privacy. I don't need big brother, I don't need you watching me at all times knowing exactly who I am from within the crowd.

Sebastien: Right.

Nathan: You know?

Sebastien: Well, so you know, and I mean, let's think about why Amazon would want to race to use or at least, set the rules on what they could apply it to. So their interest to me off the top of my head from what I know about what they're working on, and what they've already put out.

Sebastien: Number one, the first thing that comes to my mind as to why they would want to use facial recognition is perhaps their Amazon GO stores. And so that uses a different type of computer vision, that is more object recognition based to when you pick something up off the shelf, and then it charges to your Amazon account, unless you put it back.

Sebastien: So maybe facial recognition, this is somewhere they may want to apply it in those stores. And I'm going to go ahead and throw a...how do you say it? This is not a rumor, but this is a guess. I'm going to say that Whole Foods is the testing space, eventually. I believe eventually, everything that happens in Amazon GO will eventually migrate over to Whole Foods, because that's their move. Right?

Nathan: That makes sense.

Sebastien: I mean, outside of that with facial recognition you could use it for logins and authorization.

Nathan: That's what I was thinking. So like, instead of having to swipe into the store, just walk in, it knows who you are. Now when you pick things up, auto charge, no phone needed to pull up, grab what you need, dip. You don't need anything on you.

Sebastien: Here's another thing. So you know, about a year ago, they were talking about doing a big part of their strategy. Trying to get package deliveries to not sit outside, it’s a huge liability for them. So that makes sense. I don't know if you heard when they were talking about delivering packages inside of your home. So you'd buy a lock system and then Amazon will be authorized to unlock your door and their packages inside. They also talked about a similar Partnership, which would allow them to do it where they drop it off in their car. So just like you're at work, put it in your trunk. Boom. So you know what I mean? They just locate your car, unlock your trunk.

Nathan: That makes sense. That's dope actually.

Sebastien: They come to you.

Nathan: That way you don't get robbed.

Sebastien: No delivery windows. It's not sitting outside.

Nathan: Yeah.

Sebastien: And so you can imagine that facial recognition could be a way to sort of unlock the trunk or being a depending on what obviously, you know if that car has rear-view cameras. Yeah, but I believe don't quote me on this, but I believe they have investments in Ring. And that's the sort of the camera, the door cam, the security, the doorbell that has a webcam on it.

Nathan: Not Nest, Ring?

Sebastien: Correct.

Nathan: Okay.

Sebastien: Yeah. So I mean, but that's another call out right there. Right? So Google is probably going to want to jump in there. Also, Facebook is going to want to be in there with Instagram. And I mean, you know, you know, a lot of people are like, “oh, facial recognition” but then think about, for example, AR face masks. A huge part of AR face masks is facial recognition. Face ID on your phone, if you like that, that's gone.

Nathan: Yeah.

Sebastien: So there's there is some useful...

Nathan: I see what you're saying. It's like it's just hard.

Sebastien: It's all tied together. There are too many nuances. So, I mean, for me...

Nathan: There should definitely be a limitation thrown on. Because like, if you have giants like Facebook and Instagram, and who else did you mention there?

Sebastien: Google...

Nathan: Yeah, Amazon. Yeah, Google. Yeah, exactly. Like all these people using this software. I mean, that's like a lot of breaches of privacy. Yeah, so...That’s a lot of data to have on people. It's really personal.

Sebastien: Yeah. And I think that that's a big thing that like, is really important for me on the show is, I want people at home to think about, you know, you often have these issues that are these paradoxes. These things that are, you know, of course, we want AR face masks...And of course, we want sort of like the convenience of being able to use our phones and unlock things with ease without having to like, jump into an interface and click buttons. Instead, it just knows who I am and it just logs in and you forget about login right? The idea is to remove all the barriers.

Sebastien: So I think that a lot of people don't realize, that right now, the laws on facial recognition and artificial intelligence are being decided now, and our politicians should have a stance on this. I feel like they should be talking about these topics, especially as privacy becomes a big concern to our generation. And if you want to have a say in that, then you should form your opinion. You know? Which side do you fall on? Do you like the ease of technology and think it's going to, you know, help us in the long term, or is this dangerous technology that's going to end up being used against us? Is there a world where both exist? Where we have the convenience, but we also have some sort of privacy as well? Are those impossible things to hold together at the same time?

Nathan: That’s what I would want to shoot for. Because like, you don't want to like you don't want to lose out on the perks. Like, I'm not saying that Snapchat shouldn't be able to have their filters, but or that you can't unlock your phone with your face. But like, it sucks to know that you're that that data is being collected and can be used later on and like, for against you or for you. It could be either, but like, still. You know?

Sebastien: Well, so let me clear up a little bit of information that some people may be thinking that, you know, a lot of times these artificial intelligence algorithms are based on large collections of data. So it doesn't really care too much about who you are as an individual now, saying that we could hold the data and then use it in a prosecution. I don't want to say that that's not possible, but I'm saying the main use for these big companies is to get, you know, thousands or millions of faces, and then find what they all have in common. So it can track the features and add sunglasses and these kinds of fun things.

Sebastien: So it's not like they're storing, like your first and last name somewhere. And it's like, “here's his face and this is what he's thinking.” you know, it's none of that. It's more so like “what do all human faces are what all cats or what have all dogs look like?” And then based on that, you make the computer really good at recognizing it because it has all these variations, different skin colors, different, you know, face shapes and different sizes of noses. And so the computer just gets really good at finding a face. So a lot of times people are like, “Oh, it's my data. It's my privacy.” And I understand that. But I think also at the same time, you have to understand the nuance of probably what it's actually being used for.

Nathan: Mm hmm. Mm hmm. I agree.

Sebastien: But still...

[Laughing]

Nathan: Yeah. How do you know?

Sebastien: You don’t.

Nathan: How do I know that...You know...

Sebastien: You need those laws in place.

Nathan: Exactly.

Sebastien: We need those basic laws of protection, those basic laws of privacy. And it's until that we, as a people decide on what we deserve in the next generation of technology. How much privacy do we deserve? How much will we be monetized? How much? You know, there's a benefit of giving companies this data, because you get free services. You get a lot of like, ease of use, but at the same time, I agree, you can’t have some control over your data. It can't just be the wild west out here.

Nathan: And I for one think that Amazon should not be the only person figuring this out right now.

Sebastien: All right, all right, Mr. Bezos, you heard it here from Nathan. Crazy story. What y’all think?