Pod Insights Decoder Recode Decode: Rick Smith About this episode Axon CEO Rick Smith talks with Recode's Kara Swisher about his company's best known product (the Taser), how it's thinking about the ethical implications of new products aimed at police, and the controversies around facial recognition in body cameras. Plus: Is the weapon of the future a pistol that doesn't kill you?
Featuring:
Rick Smith (@AxonRick), CEO of Axon (@Axon_us) and author of The End of Killing: How Our Newest Technologies Can Solve Humanity's Oldest Problem
Hosts:
Kara Swisher (@karaswisher), Recode co-founder and editor-at-large
More to explore:
Subscribe for free to Reset, Recode's new podcast that explores why — and how — tech is changing everything.
Listen to Kara's interview with NYU Policing Project director Barry Friedman, who served on Axon's ethics board.
About Recode by Vox:
Recode by Vox helps you understand how tech is changing the world — and changing us.
Follow Us:
Newsletter: Recode Daily
Twitter: @Recode and @voxdotcom
Learn more about your ad choices. Visit podcastchoices.com/adchoices
Transcript 10,585 words · 8 speakers
0:01 Speaker 00 Company spending is complex. Managing it should be simple. This is AMEX Corporate, a single platform connecting cards, expenses, and payments to streamline the way you work. Flexible controls for when priorities shift. Support around the clock. All backed by American Express. So you can stay focused on what matters most. Built to simplify, designed to scale. This is AMEX Corporate. Get started at go.amex/ammexcorporate. Terms apply.
0:37 Speaker 01 You could try to find the perfect coffee to water ratio. Or you could press a button. With Keurig, each K-cup pod is precision measured for a perfectly delicious cup. No effort, no grind, no math required.
0:54 Speaker 01 Great coffee without the grind. Keurig.
1:00 Speaker 02 Hi, I'm Kara Swisher, editor-at-large of Recode. You may know me as someone who only likes guns when Sarah Connor is wielding them, but in my spare time, I talk tech. And you're listening to Recode Decode from the Vox Media Podcast Network. Today in the red chair is Rick Smith, the founder and CEO of Axon, the company formerly known as Taser International. In addition to Taser, Axon is best known for the body cameras it makes for police officers. And in July, it said it would not embed facial recognition technology in those cameras because of concerns over its accuracy. Rick is also the author of a book about gun control called, "The End of Killing: How Our Newest Technologies Can Solve Humanity's Oldest Problem." I'm very much looking forward to this interview. Rick, welcome to Recode Decode.
1:40 Speaker 03 Awesome. Thanks.
1:41 Speaker 02 So let's have a little bit of background about the company, about Axon. You know, everyone knows what a taser is. And taser has been deployed in different ways. But explain to me how you started the company and how it's gone to doing body cameras. And I want to sort of break down how big a business it is and where you're deployed.
1:57 Speaker 03 Yeah, so in the early 1990s, I was living in Europe. I was in graduate school. And two of my high school friends had been recently shot and killed by a businessman with a gun in the glove box. And it was a situation that just started as an argument and spun out of control. And three lives were ruined that day. My two friends are dead and the guy who killed them is spending his life in prison. I've never met him, but from everything I know, he didn't set out to kill anybody that day. He just had a gun to defend himself and got in a bad situation. So, the light bulb that went on for me is, obviously guns are a very divisive issue in this country, but for me, it seems really strange. Like, why are we still using these medieval weapons that shoot literally balls of flying metal at people?
2:41 Speaker 02 Metal.
2:41 Speaker 03 And I came to conclude, I think the way out of this is the way we get out of most real problems, that we invent our way out. And so, I started doing some research and discovered-
2:51 Speaker 02 Were you had a background in this?
2:52 Speaker 03 No, I was, well, I did have a background in biology and neurobiology in particular.
2:58 Speaker 02 Right.
2:58 Speaker 03 But I was, I hadn't had a job yet.
3:00 Speaker 02 Right. So, you were just studying in Europe?
3:02 Speaker 03 Yep.
3:02 Speaker 02 Where did this shooting take place?
3:04 Speaker 03 In Scottsdale, Arizona.
3:06 Speaker 02 Okay. So, it was after an altercation?
3:08 Speaker 03 Yep.
3:08 Speaker 02 After an altercation. So, this happened and you thought, "I'll invent something."
3:12 Speaker 03 Well, yeah. I had an entrepreneurial focus. I'd seen my dad been a serial entrepreneur. And as I was debating, you know, American gun violence with these Europeans, it just struck me that this isn't a problem our government has shown any particular aptitude at solving.
3:29 Speaker 02 Right.
3:29 Speaker 03 And ultimately, with 300 and some million firearms in the hands of the public, like, the horse has kind of left the barn.
3:34 Speaker 02 Right.
3:35 Speaker 03 Getting it back would be really hard, but if we could give something better to people, if you start to tease the logic of it all apart, the reason that people use guns to defend themselves is to stop somebody who's threatening them. And if we go even a little deeper, like, into policing, like, police are actually, they don't intentionally kill anyone, but-
3:54 Speaker 02 Well, sometimes they do. But go ahead and go ahead.
3:56 Speaker 03 Well, from at least a legal and policy perspective, the reason they use lethal force is it's the only thing that reliably stops somebody today.
4:02 Speaker 02 Sure.
4:02 Speaker 03 So, to me, that's a technology problem. If we could create Captain Kirk's phaser from Star Trek, then there'd be no reason to-
4:07 Speaker 02 Make it stun, set it to stun.
4:08 Speaker 03 Exactly.
4:09 Speaker 02 Right.
4:09 Speaker 03 And so, that's what we do.
4:11 Speaker 02 So, you had created Taser. Talk about going about creating the Taser business.
4:15 Speaker 03 So, I just started doing research and I discovered the actual Taser weapon had been around for over 20 years. It was invented by a NASA scientist back in the 60s. The early generations were pretty buggy, like many technologies. And it turned out, he lived in Arizona, near where my parents lived. And so I dialed 411, for those listeners who may not know this, before Google, that's how you'd get a phone number.
4:40 Speaker 02 911 was for the police. But go ahead.
4:42 Speaker 03 And I got his number, called him up, and next thing I know, here I am, 23 years old, standing on his doorstep, he's 73, and we just sort of fell in love over this idea that he'd been working on for a long time, and I convinced him to give it another shot.
4:59 Speaker 02 So in doing that, you wanted to create something that was debilitating but non-lethal, right, correct?
5:04 Speaker 03 Yep.
5:05 Speaker 02 The idea that you would be able to stop someone in some way, and again, the idea wasn't from Star Trek, set tasers to stun.
5:11 Speaker 03 Well, the inspiration, I'd say, came from science fiction, and then when I discovered somebody had already made some progress in this direction, that's where I went to get started.
5:21 Speaker 02 So what were the difficulties of creating something like this? Obviously, that a non-lethal thing can be lethal, I mean, if you do them wrong, they're quite lethal, and it also affects the neurobiology of something. Explain what a taser does.
5:33 Speaker 03 Yeah, so as you and I are sitting-
5:34 Speaker 02 When it started, and then what it does now.
5:36 Speaker 03 As you and I are sitting here talking today, there are numerous electrical signals in our body, all the firing of the neurons is all electrically controlled, and the idea of a taser is that we plug into the human body, and we, it's like a remote control, we flood the system with signals, and it causes your muscles to spasm. It's a little bit like an induced seizure, is probably the best analogy, but it does not require causing any physical damage, so it's really a command and control instrument. Think of it like radio jamming. We're jamming the communication so that we can impair somebody's ability in a temporary, yet reversible, way. So I've been shot with it seven times, it's certainly not fun, but the intention is, again,
6:18 Speaker 03 to stop somebody without requiring, inflicting death or permanent injury.
6:23 Speaker 02 And to replace, among police officers or people, this device, to have a taser, if you're in a, you know, I know several women who have them, to stop someone from, just disable someone, essentially.
6:36 Speaker 03 Exactly.
6:37 Speaker 02 So you started this, what's it changed from the technology of this?
6:40 Speaker 03 So the early taser generations, they were underpowered, frankly. Rodney King was shot famously with an LAPD taser, and it failed to stop him, and officers went to their batons, and the rest is catastrophic history. So one of the early things we had to do was actually go back to the lab and determine How we could tune these things so they would actually stop really focused aggressive people.
7:02 Speaker 02 Quickly, quickly.
7:03 Speaker 03 And immediately, yep.
7:04 Speaker 02 Right.
7:04 Speaker 03 And that we launched that technology in 1999, sort of the upgraded revamped taser weapons. And since then, it really took off in law enforcement where they're pretty ubiquitously used now throughout most of the English speaking world and now being adopted in Italy and France.
7:18 Speaker 02 So what is the, what are the terms of engagement, I guess, around a taser? Do people died from using tasers, presumably heart attacks?
7:25 Speaker 03 Actually, that's what is so interesting. Yes, these are not risk free, there have been deaths. There have been about 28 deaths directly caused by the taser. The biggest risk is head injuries from falls.
7:36 Speaker 02 People fall, okay.
7:37 Speaker 03 So about 6 million people have been hit with tasers, about, I think, 18 falls and eight ignitions of flammable liquids, like in a meth lab or if somebody's doused themselves in gasoline, you know, it can ignite.
7:51 Speaker 02 Because it's doing an electrical signal, correct?
7:53 Speaker 03 Yeah, it's an electric spark.
7:54 Speaker 02 Right, that's right, right.
7:55 Speaker 03 So that'll, that can set up. Now, there's a lot of debate about whether or not tasers affect the heart. Now, it's very hard to prove a negative, but in all the testing we've done, we actually see there's a very significant safety margin. In fact, the most prolific inventor of pacemakers and cardiac devices is on my board. He's taken shots directly to the chest and he and our medical advisors.
8:16 Speaker 02 You shoot people with a pacemaker? That's a risky thing.
8:19 Speaker 03 Well, pacemakers actually have to withstand by law
8:21 Speaker 02 No, I can't.
8:22 Speaker 03 The cardiac defibrillators.
8:23 Speaker 02 I still wouldn't do it.
8:24 Speaker 03 Yeah.
8:25 Speaker 02 I still wouldn't do it.
8:26 Speaker 03 It's very hard to overcome people's intuition that this electricity wouldn't
8:29 Speaker 02 It's dangerous.
8:30 Speaker 03 affect the heart.
8:31 Speaker 02 But your point is you're trying to, you're going to try to replace guns, which are absolutely deadly most, much of the time.
8:36 Speaker 03 They have about a 50% fatality rate and we're about 1 in 10,000 fatality rate.
8:41 Speaker 02 So you, you developed this thing and just from a science point of view, you're putting an electric charge into a body, right? From a distance, right? It travels, there's...
8:50 Speaker 03 So we fire two wires effectively that make the connection and then the power supply remains in the operator's hand and we're transmitting energy down the wires and into
8:59 Speaker 02 So when you click the button, I've never used a taser, so when you click the button, it tasers someone, right?
9:03 Speaker 03 Yeah, it delivers a five second burst of electricity.
9:06 Speaker 02 Right. And then people multiply tase people? I'm sorry to be so stupid, but I want people to understand.
9:11 Speaker 03 This is actually a great question because, yes, particularly, so the way to think about it, the taser is a stimulation device that is running for a set period of time. During that five seconds, somebody is impaired or incapacitated. Depending on, you want to spread the electrodes out so they fire at a spreading angle. Because the electricity, you want to actually spread it through enough nerves.
9:34 Speaker 02 Right.
9:34 Speaker 03 That it incapacitates. If you shoot somebody really close, you hit them in the forearm, it might impair the muscles in that area, but not the whole body. But even if you get a great spread and it's highly incapacitating, the moment it turns off, it's just like when I stop flexing my muscles, they're right back to normal. So particularly if police are dealing with somebody who's high on cocaine or methamphetamines, they're right back into the fight. And so sometimes police do have to use it multiple times while they're trying to get.
10:03 Speaker 02 Or use the opportunity to incapacitate them.
10:05 Speaker 03 Yeah.
10:06 Speaker 02 Correct, or cuff them presumably, right?
10:08 Speaker 03 We train, we call it cuffing under power.
10:10 Speaker 02 And talk about the training. How many people, so tasers, how wide is it on consumers versus police officers or security officers?
10:17 Speaker 03 Yeah, so overall, there's a similar number in the field. There's probably 500,000 consumer tasers and 500,000 or 600,000 that are being used, you know, by police officers.
10:28 Speaker 02 And so 500,000, 600,000 of them. So how many do you sell a year? What's the?
10:33 Speaker 03 So about 80,000 into police work. And then, you know, probably 20,000 or so into consumer.
10:42 Speaker 02 And the cost.
10:43 Speaker 03 Police tend to replace them.
10:44 Speaker 02 Yeah, they tend to replace them because they use them.
10:46 Speaker 03 Yeah.
10:47 Speaker 02 Yeah. So again, you do training for police officers when they're using these, correct?
10:51 Speaker 03 Correct.
10:51 Speaker 02 They just send them tasers and good luck, my friend.
10:54 Speaker 03 Yeah, it's, it is an area where training is really important. And we have really extended ourselves into training. Well, a lot of people don't realize, there's 18,000 police departments.
11:04 Speaker 02 Yes, sir, I do realize that, yeah.
11:05 Speaker 03 Yeah, and, you know, they have varying levels of sophistication and they may have access to varying levels of expertise. So.
11:14 Speaker 03 We've just taken on a larger role because, you know, we know this technology really well. And so we're taking on more and more of the training. Just actually today, I believe, or yesterday, we announced we're using VR now to do empathy training.
11:27 Speaker 02 Mm hmm.
11:27 Speaker 03 So turns out, police officers are called to many different situations.
11:31 Speaker 02 Sure.
11:31 Speaker 03 Including people who are suicidal or have schizophrenia. So we just did a suicide prevention experience where it's one thing for you to tell a police officer something.
11:40 Speaker 02 Right.
11:40 Speaker 03 But VR is powerful.
11:41 Speaker 02 Yeah, it's interesting. There's Jeremy Balenson at Stanford. There's all kinds of people working on this idea that you can be. My issues, there was a really interesting thing that Lorraine Bowjob backed around immigrants, where you were in a VR setting. It's a really interesting area. The question is, how can you make someone actually scared? How can you make someone actually feel? Like, can you inject them with some feeling? I was talking, you know, the idea of seeing it is better than hearing it, but that, or hearing it, you know, someone's story. But I wonder how you can actually make someone feel, you know, police officer scared on the other side of an arrest.
12:15 Speaker 03 Yeah, so, so, for example, what we did with our schizophrenia training experience. So we put the officer through a scenario where we're putting him through as the police officer, and they deal with the schizophrenic. And then we re-won it where they're in the schizophrenic's shoes. Now, of course, we can't give them schizophrenia.
12:32 Speaker 02 Right.
12:33 Speaker 03 But we can play audio video stimulus with the help of mental health experts.
12:36 Speaker 02 Right. It's just interesting.
12:38 Speaker 03 And it's sort of, it's intended, it's sort of like feeling like you're in a horror movie.
12:41 Speaker 02 Right.
12:41 Speaker 03 Where there's, there's voices and there, you're having strange perceptions. And then when you see a police officer come and point a gun at you and a flashlight, flashing lights, and they're yelling at you, you sort of suddenly understand, oh, wow, yeah, this might trigger somebody who's on the edge of their seat.
12:56 Speaker 02 So you do these as part of the trainings? Is it another product?
12:59 Speaker 03 I know, it's embedded in our TASER training.
13:02 Speaker 02 TASER training.
13:03 Speaker 03 This is relatively new.
13:03 Speaker 02 And so you've been doing TASER training for as long as you founded the company, this is the original thing. But then you moved into body cameras. Explain that. What other products do you have besides body cameras and TASER?
13:12 Speaker 03 And then we run the largest cloud software product in public safety. We have about 70. petabytes of police audio video data. So we run our body cameras and software. It's sort of like the iTunes ecosystem.
13:23 Speaker 02 Right, but they're together. They're together. That's an interesting way to put it. So talk about the getting into body cameras. When did you do that?
13:29 Speaker 03 So this started in around 2007 or 2008. And so we launched tasers around 2000 and they actually exploded in popularity because police officers found them super, super helpful.
13:42 Speaker 02 And so many comedies in Hollywood.
13:44 Speaker 03 Actually, they did.
13:45 Speaker 02 It's kind of crazy how many times I've seen them in movies and anyway.
13:49 Speaker 03 Yeah, we've become part of the pop culture lexicon. But as it became more widely used, there were legitimate concerns about overuse and abuse. And so again, you take a solutions mind to it, sat to it and say, well, we can argue about whether police are doing good things or bad things, or what if we recorded it? And so that's that simple insight got us into the body cameras to say, okay, this is it's a way to control for bad officers, hold them accountable, and protect good officers who might be falsely accused of things.
14:17 Speaker 02 Right. And so when you this idea that came up, had there been cameras on officers yet?
14:22 Speaker 03 There had been cameras in patrol cars.
14:25 Speaker 02 Right, on the deck, the deck cam.
14:27 Speaker 03 Yeah, they literally VHS tapes. So it was becoming possible that you could do wearable cameras, but nobody had really done it yet.
14:36 Speaker 02 Because there've been all kinds of wearable cameras that consumers were using.
14:39 Speaker 03 Yeah, like GoPro and all that.
14:40 Speaker 02 Right. Is that where the impetus came? Or the just that there should be this is a great new product for us to offer?
14:46 Speaker 03 I mean, it's hard to know how much inspiration we took from consumer products, etc. I think it was a combination of seeing what's happening.
14:53 Speaker 02 We're serving the police. This is something that could help them do a better job.
14:57 Speaker 03 Exactly.
14:58 Speaker 02 So talk about the early cameras, the body cams.
15:00 Speaker 03 Well, our first cameras, like many first gen products, we didn't get them right on the first attempt.
15:07 Speaker 02 And how so?
15:08 Speaker 03 They were a little too big. We built our own touchscreen computer. I'd say we perhaps got a little carried away with all the things we could engineer and it grew in complexity, size, cost. And then it was our second generation where we really focused on simplifying the product experience. And that's when they started to take off, which would have been around 2011 or 2012.
15:27 Speaker 02 All right. And how did you introduce them to police? This is what we're going to make for you?
15:32 Speaker 03 You know, there's big conferences that we go to. And then also, we would start to show them in our TASER training classes. So Taser weapons require police be trained typically annually and for instructors at least every two years. So we're putting about 10,000 police instructors a year through training and we would just, you know, showcase the cameras there as well. Hey, if you're using taser weapons, you've probably, you're aware of the public controversy about them. Here's a way to address public concern with cameras.
16:00 Speaker 02 And what was the reception of police officers at first?
16:04 Speaker 03 So with body cameras?
16:05 Speaker 02 Yeah.
16:06 Speaker 03 Not good.
16:07 Speaker 02 Yeah.
16:08 Speaker 03 As you can imagine, when you, so police do feel like they're under siege, you know, a lot of negativity in the press, et cetera. And so the initial reaction was, what, you don't trust me? Now I gotta wear a camera and you're gonna record what I do all day? What we found, though, is once they use it for a while, like within about 90 days, police will refuse to go on patrol without it because they quickly realize, oh, if I'm doing the right thing, now I'm above reproach because I have video of what I was dealing with. Whereas otherwise, everything is a he said, she said.
16:38 Speaker 02 Right, which is, it's a complaint and then, and people assume the police are lying, presumably.
16:44 Speaker 03 Yeah, well, it depends on, you know, obviously your personal point of view, but I think that the, the case that best exemplified this was the shooting of Michael Brown in Ferguson, Missouri. Right? People that believe police, believe that that officer was probably fighting for his life. People who don't believe police thought it was a murder. And we all formed our opinions based on our biases and there was very little fact.
17:04 Speaker 02 Very little fact. Okay. We're here with Rick Smith. He's the CEO of Axon and the author of The End of Killing: How Our Newest Technologies Can Solve Humanity's Oldest Problem. We're gonna take a quick break now and we'll be back after this.
18:03 Speaker 01 No effort, no grind, no settling.
18:08 Speaker 01 Great coffee without the grind. Cure it.
18:13 Speaker 02 We're here with Rick Smith. He's the CEO of Axon. And Axon, if you don't know, makes, it used to be called Taser International. In addition to tasers, it's best known for the body cameras it makes for police officers. And in July, it said it would not embed facial recognition technology in those cameras because of concerns over its accuracy. We're still talking about the cameras themselves. So you bring them to police officers, first they distrust you for putting them in, and why should we be stalked, essentially, even though they're public servants? And I would like cameras in all public facilities, but there are controversies around privacy and stuff like that. What was your argument to them about privacy? Like, well, they're public servants, so they shouldn't be private, and that when they're acting on the public's behalf in public?
18:51 Speaker 03 Yeah, as with anything, you have to make it personal to the individual. So we really focused in on, you know, look, you, as a police officer, get accused of all sorts of terrible things, and most of the time, it's not true. At least, you know, because I work with police, I'll admit, I have a bias for believing, like, that they're good people. I've met so many good people. And so for us, it was like, look, this isn't about mistrusting you. This is about giving you the tools for you to show what actually happened. And if you're doing the right thing, the truth will set you free.
19:24 Speaker 02 Right, and that, and their bad apples should be caught.
19:27 Speaker 03 Yeah, and I'll tell you, nobody hates a bad cop more than a good cop, because it gives the profession a bad name, and they want to get bad apples out of policing as much as anyone.
19:36 Speaker 02 Right, and so what these cameras do, talk about from a technological, the latest version, and how much do they cost, police, depends on how many they buy.
19:43 Speaker 03 They're like four to 800 bucks in that range.
19:45 Speaker 02 Each, and they wear them on their shoulders?
19:48 Speaker 03 Yeah, typically somewhere on the chest or shoulder.
19:50 Speaker 02 Right, okay, and they are small? Explain the device and the technology behind it.
19:55 Speaker 03 So it's, you know, I've got an iPhone Max here. It's maybe two-thirds this size, but quite a bit thicker, because we have to have a 14-hour battery, and it's recording video all day, which is a high-intensity processor, you know, a process. Now, we don't record the entire shift. It's buffering video. And then, once the officer signals it to start, we then begin a recording. We grab the previous 30 seconds of video, preserve that, until they turn it off. And then, we've now added all sorts of triggers and mechanisms. So if they pull a gun out of the holster, there's a wireless signal that turns the camera on.
20:32 Speaker 02 I see.
20:32 Speaker 03 Or if they turn on the lights and siren in the vehicle. Candidly, where I see this going, long term, we're already recording the whole day. It's just officer privacy issues that, when body cameras were new, I think there was a lot of concern. Well, we don't want to be recorded all day. We'll only record the critical incidents. But then what happens is occasionally, and especially when their adrenaline's pumping, they forget to turn the camera on. Then you're relying on sensors to turn it on. I think where we'll be in another 10 years is the camera will just store the entire day's shift. It'll upload the critical segments. But if, God forbid, they forget to turn the camera on and something bad happens, they could go back and recapture that before the upload event.
21:13 Speaker 02 So they don't have to turn it on or off themselves?
21:15 Speaker 03 Yeah. It would, well, I think they would still turn it on and off, but these would be adding markers effectively.
21:20 Speaker 02 I see. So pulling their gun out, or putting a siren on, or a call where they have to respond to something, correct? Could all set it off.
21:29 Speaker 03 Yeah, so we have sensors that do all that today. The challenge you start to get into is then, you know, officers are pulling their guns out, actually for lots of reasons. They may be going into a restaurant, they want to, or they're checking the gun, or, you know, one officer is turning on everyone else's cameras. So then sometimes they start to get annoyed that, okay, my camera's turning on all day. And so then they may start to disable some of those signals. So it's actually fairly tricky to get it to where it's only turning it on when it's really supposed to.
21:58 Speaker 02 Right, so now they only go on if officers turn them on, right? Except for the sensors.
22:01 Speaker 03 Except for the sensors, yeah.
22:03 Speaker 02 Sensors. Do you think they should be recorded all day?
22:07 Speaker 02 Enormous amount of data, we'll get into that in a second.
22:09 Speaker 03 You know, there's also what about bathroom breaks? And then there are, like, you know, these are human beings as well, they have some privacy rights, so if they're not on a call and they take a, not on, deployed in a police mission, they're in their patrol cruiser and they take a personal phone call, you know, they probably have a right to privacy, you know, not to be physically recording in a public record system, personal conversation, so it gets a little tricky.
22:32 Speaker 02 Right, right. So when you convince them to start using, how many are deployed now? Are you in all those 18,000 or?
22:37 Speaker 03 We're certainly not in all of them. We have about 350,000 cameras deployed around the world.
22:42 Speaker 02 So that make you the biggest?
22:43 Speaker 03 We are.
22:44 Speaker 02 Who else competes in that area?
22:45 Speaker 03 So Motorola just acquired our biggest competitor, which is WatchGuard. They do cameras in patrol cars and wearable cameras. There's another company called Digital Ally that's a direct competitor. Panasonic is in this space, they have in-car and body cameras, and then there's a couple others.
23:02 Speaker 02 And talk about selling to police and these kind of things. I mean, right now, they use them now, everybody uses them. Talk about the process, who buys those? Who decides, the chief or who?
23:11 Speaker 03 It depends on the agency. Like, in bigger agencies, they delegate it down. Body cameras are typically purchased more by IT people because they quickly realize managing the data is the hard part.
23:23 Speaker 02 Sure, we'll get to that.
23:24 Speaker 03 And then the taser weapons are usually bought by the same people that are buying guns.
23:28 Speaker 02 Yeah, right. And so you convince people to use them, and the idea is to protect good officers and not protect bad officers. I want to get into later the idea that citizens also have their own cameras on all the time. What's been the experience of police officers so far? They think this is a good thing or a bad thing or?
23:46 Speaker 03 Almost universally, after a few months wearing a camera, they will not go on patrol without it, so they quickly become.
23:52 Speaker 02 Because it protects them.
23:53 Speaker 03 Because it protects them.
23:54 Speaker 02 Right.
23:54 Speaker 03 I mean, police officers take more verbal abuse than about anyone except, you know, air traffic or, you know, airline reservation people when there's a storm.
24:04 Speaker 02 Right.
24:04 Speaker 03 They take a lot of abuse.
24:06 Speaker 02 Uh huh. And you, and then when you record the, you're also recording their audio and how they interact with the public.
24:11 Speaker 03 Yep.
24:12 Speaker 02 I saw an astonishing thing, it was an art project, where they took all the voices of police officers talking to people. And there was a lot of abuse stuff, it was really interesting. At the same time, they talk differently to different races. It was really interesting, there were word clouds, and it was like fascinating the difference between them talking to white people and black people, it was sort of shocking. Not surprising, but sort of weird to see in one place. Do you think it's used as sort of a learning thing of how you're talking about doing empathy with with police officers? Do you think they realize, because they're getting abuse from citizens for one, but at the same time, does it help in any way?
24:48 Speaker 03 We're in the early days there, so I mean, you're going exactly where this is headed. It starts out as, hey, this is a way to oversee officers and protect them from allegations. Then you move into more things like, well, what else can we do with this video? In a lot of cases, it really streamlines the criminal justice system. Like, for example, in domestic violence disputes in Australia, there's a great study that showed just recording victims of domestic violence the night of the event was far more impactful in gaining convictions on abusers because
25:18 Speaker 02 People get regret.
25:19 Speaker 03 When you see this person, you know, and what they've been through and they're physically hurt and they're crying, it's much more impactful than taking a written statement, you know, some days later.
25:28 Speaker 02 And then people go back on those the next morning and stuff like that.
25:30 Speaker 03 And the other one that's super interesting was the rate of continued abuse dropped. It turns out like when police would come to the call the first time, then they leave or something, you know, this is before the laws where they have to arrest the abuser on sight. Of course, the abuser's trying to intimidate the other person not to testify against them, so you may get secondary abuse calls where police, they actually learned they would do the interview that night, and then they would tell the abuser very clearly, hey, we're done, we have what we need. They're sending a message, there's no point in intimidating anyone, like, we have all the evidence that we need.
26:06 Speaker 02 Right, that we need and everything. So those are in good positions. What about the impact of all these videos? You know, there's been whole TV shows of them, of police, and it's usually the terrible things, or the weird things, like a deer jumps on the hood or something like that. When you have these, when there's so many disturbing videos where it's clear something that the police escalated it far too quickly, only because I feel the police should never escalate, like, even if citizens are being awful to them, but not violent, there should be almost no Reaction, I know that's difficult to say, I'm not a police officer. How do you think that impacts that?
26:44 Speaker 03 Well, it's a bit tricky because even if police are doing the right thing, it's still not
26:50 Speaker 02 Pretty.
26:51 Speaker 03 It's not pretty. I mean,
26:52 Speaker 02 Right.
26:52 Speaker 03 When police are using force on someone, it's disturbing, to say the least. And that's if it's fully justified. And if it's not justified, like, you know, police are humans too, and sometimes they're making mistakes, or they're escalating things up. So it's, I think overall, it's very helpful because it's giving people context of the challenges police are dealing with. And we are very much focused, over the next five to 10 years, in taking the video feed, and now using it for training, and feeding it back to officers, and helping them, you know, learn from how they're interacting. And looking at issues like, you know, we talked about racial bias and other things like, I think even the critics in this space would say a lot of it is unconscious bias,
27:35 Speaker 03 that, you know, officers don't self-identify as racists. But, you know, these things are happening. And so the degree that you can use these as training tools to come back and help officers identify and learn, and know they're going to be held accountable. I think there's just so many opportunities to continue to improve the system.
27:51 Speaker 02 Do you think being watched minimizes bad behavior?
27:54 Speaker 03 100%. Yeah, I mean, the observer effect, in general, has been pretty well documented, that human beings who are being observed, or who even believe they're being observed, behave better.
28:05 Speaker 02 So why the spate of sort of really disturbing videos? You know, the ones at the car stops where it's clear the officer escalated it? There's been lots of those, it seems like. Or they said it happened this way, and then the video shows completely the opposite.
28:20 Speaker 03 Well, you know, there's 800,000 police officers. And with 350,000 of them wearing cameras, you are going to have some where, you know, maybe that officer's having a bad day, and they lose it, or just bad judgment. So I tend to look at it as sort of a glass half-full, that, you know, given the huge number, there's a relatively small number of videos that go out of control, but those tend to get all the attention, right? We don't see the 300,000 videos a day where officers are doing the right thing. You see the few where it goes off the rails.
28:53 Speaker 02 They could put those up.
28:55 Speaker 03 Yeah, but you know, who's gonna watch them?
28:57 Speaker 02 Lots of people. People watch Ring videos, but it's always people stealing your packages, which is interesting. We'll talk about that in a minute. So I want to get into, in the next section, talking about facial recognition. I just want to say clearly, you don't want facial recognition techno. I just interviewed Andy Jassy from Amazon, which makes recognition, which is a facial recognition program. There's lots of them around. You do not want them in these cameras.
29:19 Speaker 03 Well, I think it's a little more nuanced than that.
29:21 Speaker 02 All right.
29:22 Speaker 03 I think it is a complicated and powerful technology that has risks of abuse and risks of inaccuracy, and our approach has been, look, if we look at it right now, the accuracy levels that we could deliver are not high enough. And the racial disparities, that is an explosive problem. If you get it wrong, and police escalate and use force on somebody, and God forbid somebody gets hurt or killed, that is catastrophic. So that risk alone has made us say, we're gonna hold off, the risk of a bad outcome is not worth it today. Then there are also, so we formed an AI ethics board to help us think this stuff through.
30:04 Speaker 02 Right, you did. Explain that.
30:05 Speaker 03 Well, basically, everything we do can be abused. Our company in particular.
30:09 Speaker 02 Right.
30:10 Speaker 03 Taser weapons, cameras, data.
30:11 Speaker 02 Right.
30:12 Speaker 03 There's privacy issues, and, look, we want to be proud of the work we're doing, and we think that getting it right pays big dividends. Police ultimately are accountable to the public and to the elected officials. So in the case of face recognition, there's the accuracy issues, there's the bias issues, and then there are legitimate privacy and search concerns. Like, under what circumstances should police be searching people's faces against the database? And I think most people would agree, if you're in an airport, and they're searching against the terrorist wanted list, everybody agrees. If they're at a protest, and they're scanning and logging everybody's face and retaining that data in a government database, like, everybody agrees that's a bad use. And then, there's a whole bunch
30:55 Speaker 03 in the middle. And I think Brad Smith from Microsoft just wrote his book, Tools and Weapons, about the need for some regulatory guidelines. And I think this is one where, you know, how do you interpret the Fourth Amendment when this technology was unimaginable?
31:09 Speaker 02 On office search. Yeah, absolutely.
31:10 Speaker 03 And so, in our case, I'm not saying that never. I could envision, when, if face recognition gets to 100% accuracy,
31:18 Speaker 03 Then it probably do a better job than a police officer in helping them understand the situation, so it could be a very valuable tool.
31:24 Speaker 02 Right.
31:24 Speaker 03 But we also just need to then understand, what is the right level of surveillance within society? I don't believe it's zero, but I don't believe it's pervasive. We don't have great answers to that yet, and so we would rather take our time and get it right.
31:37 Speaker 02 You do realize this is unusual and mature on your part. I just, like, a lot of the people I interview are like, "Yeah, we're just gonna roll it out." I mean, Andy and I had this back and forth about, you know, we're not responsible for the platform, and even if there are mistakes, I wouldn't put out a product that had not a, in that area, you know, not being worried about that issue of misidentifying people.
31:57 Speaker 03 I think we're more sensitive to it, because every product we do is steeped in these problems. When we launched the first Taser weapon, for example, there was concerns about, well, what if criminals start using these? So rather than say, "That's not my problem," we looked at that and said, "Well, that is our problem, because it would be catastrophic for us and society." And look, I don't want to be waking up in the middle of the night, reading about it.
32:17 Speaker 02 That's a cop, yeah. Like, a criminal saying, "That's a cop, that's an undercover cop," or whatever.
32:21 Speaker 03 So, we, we basically, we started serializing our ammunition. When you shoot a Taser, it sprays the area with confetti printed with your serial number. That's so we can know who shot it and hold them accountable. And I think we have learned that doing the right thing, this is gonna sound trite, but doing the right thing pays long-term dividends. And in our business, we can't afford to get it wrong. And I respect Amazon, I, I love the company, I'm a customer. You know, most of what they do is pretty innocuous stuff, and so this is, I think, a little bit of a newer area for them. And in most tech, it's like, hey, caveat emptor. In the public safety space, I tend to take a view, just from our own experience, that it pays to be careful and take your time of that.
33:02 Speaker 02 Okay. We're here with Rick Smith. He's the CEO of Axon, which is the maker of body cameras that police use, and also Tasers. He's also the author of a book called The End of Killing: How Our Newest Technology Can Solve Humanity's Problem. We're gonna talk about that book, and also some issues he's had with Amazon, not just around the facial recognition software, when we get back.
33:20 Speaker 04 Priceline Negotiator.
33:22 Speaker 05 It's me, the Priceline Negotiator. We don't need the jingle twice.
33:27 Speaker 04 What about a third time?
33:29 Speaker 05 Stop it. This is about vacation inflation and how Priceline negotiates amazing deals on hotels, flights, and rental cars.
33:35 Speaker 04 Seems like they just started.
33:37 Speaker 05 Yeah, but I didn't mention that you can save up to 60% off hotels in the Priceline app.
33:42 Speaker 04 Time to read the tagline.
33:43 Speaker 05 Fine. No one deals more deals than...
33:45 Speaker 04 Priceline.
33:46 Speaker 05 Please stop.
33:47 Speaker 04 Priceline.
33:48 Speaker 05 Touche.
33:49 Speaker 04 Priceline.
33:50 Speaker 05 Priceline.
33:51 Speaker 06 How about a latte?
33:52 Speaker 01 Maybe dark roast?
33:53 Speaker 06 Vanilla?
33:54 Speaker 01 Caramel?
33:55 Speaker 06 Blueberry?
33:55 Speaker 01 You could get them to agree on one flavor or give everyone what they want. With Keurig, there's over 500 K-Cup high varieties of fresh, flavorful coffee. No effort, no grind, no settling. Great coffee without the grind. Keurig.
34:18 Speaker 02 We're here with Rick Smith, the CEO of Axon and the author of The End of Killing. Axon makes body cameras, tasers, for police officers and others that I never thought of criminals using. I can see anybody using a taser, but not a body camera. Of course, if facial recognition was in there, it would be very useful for criminals to use those or to identify different people. But let's stick to police officers. So you said somewhere between zero surveillance, zero and 100%. Like, there's somewhere in there. Where do you imagine it's going? Because it seems like facial recognition is going to be rolled out everywhere. Obviously, in China, it's sort of the dystopian worst-case scenario, but it's not unknown here. There's cameras everywhere in this country.
34:58 Speaker 03 Yeah, so if we look at the arc of history, one of my favorite authors right now is Steven Pinker, who just booked Better Angels of Our Nature. He documents sort of the decline in violence in human society. And a lot of that is because, look, we delegated to the government the use of force on our behalf, for the most part.
35:14 Speaker 02 Right.
35:14 Speaker 03 I mean, in the old days, you used to carry a sword and you'd fight with anybody who looked at you the wrong way. So the right level of surveillance, I think, is... Again, it's complicated. I don't think we can go back to the 1950s. There's cameras everywhere. You and I have them on our phones right now. Businesses are putting them up. People are putting them on their doorbells. I don't think we can go back. So I think the the simple answers-
35:35 Speaker 02 We don't think we will go back. We could, but we won't.
35:37 Speaker 03 Yeah, you know.
35:38 Speaker 02 Yeah.
35:39 Speaker 03 I would even go, I don't think we could.
35:40 Speaker 02 Right.
35:41 Speaker 03 But I would say the the challenge then is, okay, there's a ton of data. The answer is not going to fit on a bumper sticker, like, yes, cameras, no cameras. It's much more nuanced. How do we handle the data? How do we make sure? Like, in my neighborhood, if we're using video to capture violent criminals or even people are stealing things, great. If we're using it to discriminate when somebody, you know, is of a different ethnicity shows up in the neighborhood, that's terrible.
36:07 Speaker 02 Ring. There's a comp complaints around Ring doing that.
36:10 Speaker 03 So, you know-
36:12 Speaker 02 Or next door.
36:12 Speaker 03 I think there's going to be a lot of discussion about data privacy, data security. In particular, I think for police, you want really robust logs around who's searching this? When are they searching it? What are their justifications? You're going to want good oversight so that it's not personal vendettas. You hear stories of, like, a police officer running vehicle searches on his ex-wife's boyfriend.
36:34 Speaker 02 Yep, yep, yep.
36:34 Speaker 03 You know.
36:35 Speaker 02 They did that at Facebook, too, in the early days. There was that happened because the data was available.
36:41 Speaker 03 Yeah, so I think putting solid data controls in place, good audit functionality. And, for example, things like two key systems where an administrator needs to approve something or, in certain cases, a court, where you get a court order to do certain types of searches. What's interesting is today, a reason a lot of that doesn't happen is it's just the bureaucracy of government. Getting a warrant means you've got to go, an officer's got to go spend hours at a court with a piece of paper. So that's an area-
37:06 Speaker 02 That's the point, but go ahead.
37:07 Speaker 03 Well, that's, but the the practical effect of that is-
37:12 Speaker 02 Got it.
37:12 Speaker 03 Officers find reasons not to do it.
37:14 Speaker 02 Yes, they go around.
37:15 Speaker 03 So we're looking, actually Barry Friedman, who's on my ethics board-
37:19 Speaker 02 I just did a podcast with him.
37:20 Speaker 03 Oh, you did? I love Barry.
37:21 Speaker 02 Yeah, he's great. He's at the NYU Policing Project at NYU.
37:25 Speaker 03 It was great. He brings a very different perspective than I do. He's a police oversight guy.
37:29 Speaker 02 Yeah.
37:30 Speaker 03 But he really educated me on the Fourth Amendment and how to think about this more differently. Because obviously, being around cops all the time, I tend to hear about how everything's just painful and they can't do their job.
37:39 Speaker 02 They're so good all the time, yeah.
37:41 Speaker 03 And so at this, okay, well geez, what if we made it really easy for you to send a warrant to the court? And Barry's point was like, that would be fantastic, because just the act of asking someone makes people more thoughtful.
37:52 Speaker 02 Sure, absolutely. So talk about this board, because Barry Friedman is someone who is, I wouldn't say is anti-cop, but he's certainly wary of, and like wary of the police powers. I would say that's probably accurate. And is watching that once police seem to be better, essentially. And he operates in NYU in New York, where there's been lots of controversies around the police department there over many years. Talk about this AI board. Why'd you decide to do this? The fact that you're using the word nuance is fantastic, as far as I'm, because I never hear it from tech people, so it's a pleasure. But the AI board that you want to do, a lot of people think, okay, like, Facebook just announced a Supreme Court of Facebook that's going to watch things. Everyone thinks it's a PR thing. Talk about why you did this.
38:30 Speaker 03 So, all right, so.
38:32 Speaker 02 Because if you have Barry Friedman, it's not a PR thing, that I can tell.
38:35 Speaker 03 It's not. Now, I would say maybe it started as a bit of a PR reaction. So here's what happened. So we have all these body cameras and all this video, and we realized, wow, there's a lot of value we could help create from this video. So we went out and we acquired two small companies with AI teams to help us figure this out. And as soon as we did, bang, like, the first week, we had all this negative media about how we were going to do facial recognition, and we were going to turn the US into China. And candidly, we were like, so wait a minute, who said anything about face recognition? We actually were, when we bought these companies, the first problem we wanted them to work on was redacting faces out of videos, because our customers have to do that by hand, frame by frame, super expensive, solving expensive problems for our customers.
39:19 Speaker 02 So just average citizens shouldn't be in these videos, right?
39:22 Speaker 03 Yeah, if you're going to release a video, you have to protect the.
39:24 Speaker 02 The people in the, yeah.
39:25 Speaker 03 The idea of the people in it.
39:26 Speaker 02 Right.
39:27 Speaker 03 But seeing that.
39:29 Speaker 02 These are bystanders, right?
39:30 Speaker 03 Bystanders or victims.
39:32 Speaker 02 Right.
39:32 Speaker 03 Children, you know, children, etc.
39:34 Speaker 02 Right.
39:35 Speaker 03 Which, by the way, you go down a whole rat hole. When you talk about like, police releasing every video, it sounds great, but then there's so much nuance. Well, they arrive to a sex crime scene involving a child.
39:45 Speaker 02 Right.
39:45 Speaker 03 You want to release that video.
39:46 Speaker 02 Right.
39:46 Speaker 03 Certainly, you want to redact all the ideas, etc.
39:48 Speaker 02 You talked about that, the idea that it takes a minute to release things. You have to, I was like, why don't they release them immediately? It's like, well, you do have, you know, the prosecutors have to build a case, and so you can't immediately release, release any video.
39:58 Speaker 03 Yeah, well, and the-
39:59 Speaker 02 Even though the public pressure is on to do so now.
40:01 Speaker 03 Well, or if a cop's in your house. Do you want a police officer releasing a video of everything inside your house? What about your privacy rights? What about the people that are involved? So, it's, it's nuanced, it's complicated. Alright, so, we're facing a backlash over face recognition. And at the time, we said, it wasn't even on our roadmap to be doing face recognition.
40:17 Speaker 02 Right.
40:18 Speaker 03 Sure, we'd thought about it, but-
40:19 Speaker 02 Sure, you have to.
40:20 Speaker 03 So, punchline was, I was talking to a couple of different people, including Ragini Dugan, who used to run DARPA.
40:26 Speaker 02 Mm-hmm. Worked at Facebook and Google.
40:28 Speaker 03 Yep, I was, candidly, I was trying to recruit her to come on our board unsuccessfully. She's very busy. But, in that process, as we were talking about it, she said, "Hey, you look what they did with the Human Genome Project. They created an oversight board."
40:40 Speaker 02 Mm-hmm.
40:40 Speaker 03 Because, sure, there were lots of ways the Human Genome Project could go wrong, so investing in oversight. And, based on that idea, we created this ethics board. And I can't tell you, it's really been fun and productive. Because, look, policing is a divisive issue. Most of the discourse ends up quite polarized. You've got Black Lives Matter and Blue Lives Matter, and you've got a lot of, like, emotion. And it's hard to solve problems across that divide. So, with our ethics board, we focused on really, it's predominantly civil libertarians, although we do have a few police representatives as well.
41:14 Speaker 02 You'd have to have them.
41:15 Speaker 03 Yeah, you gotta have that interchange.
41:16 Speaker 02 Right.
41:17 Speaker 03 And- Frankly, it is, I just came yesterday, or two days ago, from our latest board meeting.
41:23 Speaker 02 How often do you have them?
41:24 Speaker 03 I think about twice a year. It's a little dynamic as to how often we need them. But this is turning into a major competitive advantage for us. Because I have come to so much more deeply understand the position of the critics. Because now I'm no longer talking to cops about the critics. I'm talking to the critics. And when Barry says, "Oh, here's my problem with the Fourth Amendment in search..." Like, I can go deep on that, and he's challenging my thinking. And Barry is what I would consider the, a true liberal. Meaning, open to new ideas, and challenging, based on the merits of the conversation. So if you stay tuned, in the next month or two, we'll be making announcements where we're... So face recognition was pretty straightforward. We're not going to do it, it's not ready.
42:06 Speaker 02 Right.
42:07 Speaker 03 But I think we're actually finding ways that we can use ethical design as a competitive advantage.
42:12 Speaker 02 Right.
42:13 Speaker 03 And I think our board's really excited about this, too. Saying, "Wow, this is, we're going to have an impact."
42:17 Speaker 02 That's a big topic, ethical design. I always say, I, when I'm talking to product engineers and things like that, I'm like, I make a joke all the time, I'm like, "Imagine it's an episode of Black Mirror, and then don't make it, or change it." So it's not an episode of Black Mirror. You know what I mean?
42:30 Speaker 03 Yeah.
42:30 Speaker 02 Think of the worst thing that could happen with this technology. And think hard, and think how easy or hard it is for that, to get to that outcome. And I think the idea of consequences is lost on so many people in technology. Or thinking of the ethical implications of things. The sort of move fast and break things thing, even though Facebook has removed that from there, for example. It's still the ethos throughout technology. You know, and then worry about it later.
42:53 Speaker 03 Yeah, I think in our case, you know, I'm a founder entrepreneur, this is my baby. I don't want to get to the end of my career like Mr. Nobel, and have to create a prize to undo the damage.
43:03 Speaker 02 Right, right. Well, I don't think you're going to get into the Belgian Congo thing, but yes, but I get your point.
43:08 Speaker 03 I mean, we would...
43:08 Speaker 02 Why is that ethos not in it? You have tech people on your board now, are they open to that?
43:13 Speaker 03 Oh, 100%.
43:14 Speaker 02 Yeah, you have Hadi, who else? Hadi Partovi.
43:16 Speaker 03 So we had Brett Taylor, he recently retired, when he became president of Salesforce, got too busy. We just added Caitlin Kalanowski, head of VR at Oculus and Facebook. The former Surgeon General, Rich Carmona. Mark Kroll, former CTO of a medical company. We've got Julie Kullman.
43:32 Speaker 02 You're saying it's a competitive advantage, but you, as an entrepreneur, how does that make you feel? This is an easy one.
43:39 Speaker 03 I love it. Because... All right, when we do things that are controversial, because of this ethics board, I have a pretty good understanding of where the critiques are going to come from.
43:49 Speaker 02 So you know, at least you know.
43:50 Speaker 03 And guess what? We actually, if we can turn those critics into advocates for our system, because they're going to say, look, if you're going to deploy AI technology X, you need to be really careful about these things. And when we launch it, we can say, guess what? We've built controls on A, B, C, and D.
44:08 Speaker 02 Right, you've anticipated as much as you could. You can anticipate everything, for sure.
44:11 Speaker 03 And ultimately, look, police, again, they report to elected councils. I'm in this to build a business over the next 10, 20, 30 years. And understanding those issues upfront allows us to be more competitive and a sustainable environment.
44:26 Speaker 02 How can we get more tech people to think like that? Like, the idea of, one, nuance, and, two, anticipation of consequences?
44:31 Speaker 03 I don't know, I could get on a soapbox and tell how other people do. In our case, it's just because the influence, we're so focused in this area.
44:39 Speaker 02 Right.
44:39 Speaker 03 It's just very...
44:40 Speaker 02 Well, you're in a more dangerous zone, that you're easier to anticipate the dangers. It's easier. So when do you imagine you can use facial recognition?
44:46 Speaker 03 I'd say three to five years.
44:49 Speaker 02 So you'll be testing it to see, to get 100? But 100% accuracy is what you do.
44:53 Speaker 03 Well, 100% is a pretty high threshold. I think, let me give you an example where we may use it at lower accuracy. Let's say we can upload lists of people that are voluntarily added by their families. People with mental disabilities, or, you know, Alzheimer's, et cetera.
45:08 Speaker 02 And they have the ability to do that? They have the right to do that?
45:11 Speaker 03 Yeah, so if people opt in, and police are using it, in that case, if I get a false match, it's no big deal. But it might prevent me from, oh, wait a minute, this guy who's acting weird and kind of violent, oh, he's got a mental disorder. We could shift him in the right direction. We think that's a, you could accept lower accuracy thresholds than if you, now, let's say we're scanning against wanted databases where it may escalate things. I think there's a different accuracy threshold there.
45:35 Speaker 02 Opt in, people opt in to the being...
45:37 Speaker 03 Opt in, and what is the consequence of error? If the consequence of error is too high, then you're going to have a higher accuracy threshold.
45:44 Speaker 02 And then lastly, using, I want to get to the Amazon thing before we finish, putting driver's license photos in these things, using them as a data set. That's something that nobody agreed to when they took their driver's license photo. How do you look at that?
46:00 Speaker 03 Some of it depends on how is it being used.
46:03 Speaker 02 Right.
46:04 Speaker 03 Like if you're scanning somebody for identification purposes, you know, or they don't have their ID.
46:10 Speaker 02 Right.
46:10 Speaker 03 If you're using it to identify them, once police have made a lawful stop and they're entitled to inquire somebody's ID, I think that's a different threshold than somebody who's just walking by a police officer.
46:19 Speaker 02 Right. It's a really problematic thing. It's just you, the issue is they often don't say, and that's, I think they have, if they want to use it, they need to say. And then we all have to agree to that being okay, or some election where we say, okay, that's going to be okay.
46:32 Speaker 03 I think that's berries. The primary thing I've taken away from him is, look, if the democratically elected officials authorize it, great, but police shouldn't be just like doing what they can because it's possible.
46:44 Speaker 02 Because data centers are so fantastic. What do you, lastly, what do you think about the cities banning these things, banning cameras or the idea, or facial recognition, not cameras?
46:54 Speaker 03 Well, California, I think, has gotten to a better and more nuanced place. So they started with a ban. I think that's a bad idea just because...
47:02 Speaker 02 This is go out of cameras.
47:03 Speaker 03 Well, yeah, but over time, like, look, in another 10 years, overturning laws is really, really hard. And so we may come up with lots of use cases, like the Alzheimer's one I shared, where police would use it and everybody agrees, but now you'd have this law that blocks you. So California's moved towards a moratorium, which I think is a prudent approach.
47:22 Speaker 02 A prudent approach. Okay. Do you see a lot that, more of that, coming from states?
47:26 Speaker 03 Well, it's hard to say. You know, to be honest, our position, most, most of the big cities are using our platform, and so we've kind of instituted a private ban, including San Francisco, LA, all the big cities in California, at least the biggest ones. What we're trying to do, part of this too, the way we can bring our customers along is say, look, you want us to be careful. Otherwise, you're going to get this stuff banned.
47:50 Speaker 02 Right.
47:50 Speaker 03 And then we can't use it even for the good use cases.
47:52 Speaker 02 Right, absolutely. Lastly, you're an author of this book, Gun, about Gun Control. You had sent a clip of Amazon blocking the books. Talk really quickly about the end of killing.
48:03 Speaker 03 Yeah, so it's, it's actually not about gun control, but it's about how do we, if we want to solve the violence problem, we need to basically invent better things than guns, so people choose to go that direction and we reduce the carnage. And one of the things that's challenging is, look, a lot of the tech companies will do things that are well-intentioned. And one of them that affects me directly is they ban ads for sensitive topics. So Amazon puts this ban in place in response to all the gun violence. They say, "We're going to ban ads with pictures of guns on it." Well, I've got a book about how to end gun violence. It's got a picture of a gun on it, and they blocked my ad. And my point is, like, that's actually counterproductive.
48:44 Speaker 02 Right.
48:45 Speaker 03 And we're not going to stop people from buying guns. They're ubiquitous. People know what they are. But when we invent something like a taser that's an alternative to a gun, we need to educate the public. And when Google and Facebook block our, and Amazon, block our ads, they actually block our ability to move the market in different directions. So they're actually locking in the status quo, and they intend to make gun violence better, but I think they're reducing our ability to change it.
49:12 Speaker 02 Yeah, they don't do nuance well, Rick. I don't know if you know that. It's happened a lot with lots of ads, where there, you know, pictures of breasts. When they were trying to do breast cancer stuff, they were blocking everything. You know what I mean? It's an interesting thing. So what would you want from them? To be nuanced, to really pay attention to what they're blocking precisely?
49:28 Speaker 03 Yeah, you know, I think they're well-intentioned, but ultimately, it's really hard to, and this is where censorship and free speech comes in. It's just so hard to have a blanket rule about what you censor and what you don't. But in a minute, I think they ought to have, like, what is the intention of this? And that they ought to have an appeals process. We've tried to get appeals, and we just get blocked.
49:46 Speaker 02 But.
49:46 Speaker 03 Where they say, "Nope, your weapon." "Oh, wait a minute, we're a weapon. It's far less dangerous, and we're trying to displace deadly weapons." "Nope, policy says, you know, no weapons, no self-defense tools. Sorry." And it's basically cut off, so we can't do any advertising on the online platforms, which I think slows the pace of change.
50:03 Speaker 02 Right, absolutely. Last question, Rick. This has been fantastic. What do you imagine, after taser, what is the, we're going to be doing a lot on technology of guns and the things where it won't go off unless the person has it, and different things like that. How do you see weapons? You talked about we, it's medieval to send a ball of metal into someone's heart, right? It's just medieval. What would be the weapon of the future for you?
50:27 Speaker 03 So, part of the reason I wrote the book was to rally my customers, my engineers, around our moonshot. And that is, within 10 years, our goal, and we will do it, is to launch a pistol. That doesn't kill you, something that's more effective than a nine millimeter at doing what you want it to do, which is stop the person, and yet it doesn't leave a permanent injury or kill him. And when we do that, that's going to be a game changer. I'm starting, and it's tricky. Like, when you talk to a police officer, the last thing they want to hear is, "You're going to take my gun." And I have to be real careful. That's not what we're about. We're going to give you something else, because when you go through that door and you have a gun in your hand, that means you've got to make a life or death decision. Boom, less than a second, and that is going to impact the rest of your life, and you can't take it back.
51:12 Speaker 03 Oops, that was a 13-year-old kid with a cellphone, not a guy with a gun. And I'm finding it's resonating with customers where their response to me is this, "Okay, but I'm skeptical. You're going to have to prove that you can deliver something that's as reliable as the thing I'm slinging on my hip right now."
51:27 Speaker 02 Right.
51:28 Speaker 03 But pretty universally, what I'm hearing is, "If you do it, then we'll use that instead." And that's, I think, going to be a huge step forward.
51:35 Speaker 02 That would be fascinating. I hope you can invent that. That would be wonderful, and many tech people should help you do so, because I think it's the carnage is just so ridiculous at this point, and it's sad. It's such an indictment on our country that we don't think of things more innovatively going forward. Rick, thank you so much for coming on the show. You can follow me on Twitter at Kara Swisher. My executive producer, Erica Anderson, is at Erica America. My producer, Eric Johnson, is at HeyHeyESJ. Rick, where can people find you and your products online?
52:04 Speaker 03 So, for the company, it's axon.com, A-X-O-N dot com, or for me, same as my book, endofkilling.com.
52:11 Speaker 02 Okay. If you liked this episode, we'd really appreciate it if you shared it with a friend, and make sure to check out our other podcasts, Recode Media, Pivot, and Land of the Giants. Just search for them on your podcasting app of choice. Thanks also to our editor, Joel Robey. Thanks for listening to this episode of Recode Decode. I'll be back here on Monday. Tune in then.
53:04 Speaker 07 20th Century Studios presents Whale Fall. The most incredible survival thriller of the year. I need the whale! Help me! In theaters Friday. To survive the unthinkable, do the impossible. You're gonna become so dangerous that the predators can't help but let you go. Austin Abrams. I'm going home! Josh Doblin. Survivor, survivor. Whale Fall. Rated PG-13. Maybe inappropriate for children under 13. In theaters Friday. Get tickets now.
Previous Recode Decode: Barry Friedman Next Recode Decode: Deborah Rutter Page updated October 13, 2026 · transcript and summary are machine-generated