How to Stop Technology from Destabilizing the World | Salesforce

Good morning, everyone. How are we? Good. Good. Listen, I am super excited to introduce the first keynote today. Tristan Harris from the center for Humane Technology. There is a reason why this is the first keynote of the day here for Dreamforce. Now, often when we think about technology and how it affects us, how we use it, we think about technology as a tool that we have some control over in different ways. But what Tristan and his center and his work have really highlighted is the complex, almost co evolutionary relationship that we as humans have with tech and how important it is the more powerful that technology becomes for us to be incredibly thoughtful and deliberate about what we're doing with the design of technology, with the business models we deploy, and with the way we understand how people and groups in society are affected by it. Now, Tristan is going to take this away for you, but I just want to give you three quick insights into who Tristan is to give you a bit of framing and context for the next half an hour. So the first, you may have heard that Tristan was a child magician. As a kid, he learned from an early age the art of persuasion, by which we often mean manipulation and misdirection. So from an early age, he's been acting in the way that he'll explain. Technology often acts on us. The second thing is Tristan studied at Stanford in the technology persuasion lab alongside the the founders of Instagram, who you may have read, just left Facebook recently. And the third thing about Tristan is he was Google's first design ethicist. So here we have someone today who has the ability from an early age, as well as throughout his career in technology, to explain and understand the deep threats, the misleading myths, but also the incredible opportunities that technology has to offer. And with that, I please welcome Tristan Harris from the center for Humane Technology.

Thank you. Thank you all. It's so exciting for me to be here. And thank you so much for choosing to come here. I appreciate it. So today I want to talk about how technology is destabilizing the world and more importantly, what we can do about that when we talk about technology. So make sure we got here. So what I want to talk about is how technology is doing that and give you one model instead of this generic idea that there's tech backlash. We were for tech and now we're away from tech and we just got to get rid of it. That's not what this is about. We need to do technology differently, but we need to have an accurate diagnosis of what's happening. And to help make that diagnosis, I want to show you a quick video thank you.

I check my phone a lot more.

Than I can count.

I check it usually when I wake up in the morning, playing a game, liking pictures. I'm constantly checking my phone until I go to bed. I think it's a mixture of everything that's on social media that makes it so cool. Stories or Bitmojis or gifts. Teens want to try out the new feature. They want to be on the cutting edge of the technology that's coming out. There is a lot of pressure to present a version of yourself that's close to perfect.

If I'm doing my homework on my laptop, I might be intermittently checking my phone. Sometimes if I just get too bored of my homework, I might play Candy Crush or something. I feel like I'm distracted a lot. I procrastinate.

I have definitely been emotionally affected by social media, maybe even depressed, but it's hard to say which causes what. Sometimes.

I sometimes wish that it wasn't a thing, so I could just hang out with my friends and play basketball or do other things like that. Okay, so this video describes one aspect and gives you that felt sense, that taste of what we all know is happening, especially with the lives of children. Children. But what I want to talk about, I think when we think about what's driving global forces in the world, like, where is the world going? People think about big oil. They think about financial markets, and we don't always think, is technology driving where.

The world is going.

And so what I want to talk to you about today is convince you that technology is its own force that's pushing culture in a certain direction. And we can predict what that direction is, and we can steer it differently. So this video was about persuasion. When you think of technology, you think, oh, kids these days, they're just using it like we used to gossip on the telephone. Adults always worry about technology. There's nothing to worry about here. But what I want to convince you is that that's not true. Because as Nick mentioned, technology is persuasive. I know this because that's me when I was a kid. I was a child magician. So it was my mother's 50th birthday party. And I learned at an early age that magic teaches you to see human minds a different way. Instead of looking at what's happening in people's choices and how they make sense of the world and their beliefs as authoritative, you start to see that their attention can be manipulated, their choices can be manipulated, their. Their notion of cause and effect can be manipulated. And so you see the world Kind of inside out. And later, as Nick mentioned, I studied at a lab called the Stanford Persuasive Technology Lab that teaches engineering students how to apply everything we know about persuasion to technology. Could persuasion be applied to the way technology drives people's attitudes, beliefs, and behaviors? And I studied with the founders of Instagram in that class. They're friends of mine. And we learned all of these techniques from clicker training, from dogs and casinos and how casinos are designed and slot machines are designed, to Edward Bernays and Robert Cialdini, things that you might have heard of in marketing. And what I learned is that technology is a very powerful force for persuading what people do. And more specifically, it's always going to be used for this. Because there's one thing that's not going away, and that's the race to capture human attention. There's only so much attention out there, and we have to capture it. Whether you're a politician or a businessman or an app or a conference or a friend. We have to fit all of this stuff in our lives into the finite amount of human attention there is. It takes nine months to grow new human attention. And so it becomes this race to the bottom of the brainstem for who's going to get attention by going lower on the psychological persuasion stack than someone else. So let me show you an example of how that works. Let's say your friend sends you a YouTube video, and right before you open the video, you think to yourself, you know, I know those other times I get sucked into YouTube and I wake up two hours later from a trance. But this time, no, this time, really, it's going to be different. And then of course, you watch the video and you wake up and it's two hours later, and it's two in the morning, and your face is in your phone and you're like, what the hell just happened? And it's. Because what you didn't realize is that there was a supercomputer pointed at your brain. We're not just persuading people by auto playing the next video on a countdown, which, by the way, is a magician's trick. You take away something called a stopping cue. When you're drinking a glass of wine. How do you know when to stop? It's a. Because you finished at the bottom. And then your mind has to wake up and it's a stopping cue. And you say, do I want some more? On YouTube, what you do is you take away the stopping cue. So you make it autoplay. 5, 4, 3, 2, 1. And I know the product manager at YouTube who designed it that way. And that's what gets people to watch more. But it's not just enough that we play these little tricks to get people's attention and get you to watch one more video. We need to point supercomputers at your brain now, because in the arms raised for attention, that's where we've gotten. And what happens when humans play against supercomputers? We lose. Think about what happens when Garry Kasparov plays a regular, plays you or me at chess. He can just see more moves ahead on the chessboard than you can. So no matter what move you play, he always wins because he just can see more moves ahead. So when you land at YouTube, you think you're going to watch one video, but you're activating a supercomputer that's learned based on what 2 billion human animals have ever clicked on. What's the perfect thing I can play to keep you here next? And it's so good now that 70% of what YouTube views and traffic is now based on the recommendations that an algorithm is providing from the chief product officer at YouTube. YouTube's average watch time is now 60 minutes because of what our recommendation engines are putting in front of you. So technology. If you have 2 billion people jacked into this system, you now have this kind of digital Frankenstein who is basically steering the whole world's behavior. Because there's 1.9 billion people who use YouTube. That's about the number of followers of Islam. So if you compare sources of psychological influence and daily psychological influence, there's 2.2 billion users of Facebook. That's about the size of the followers of Christianity. Except there's no one home. Increasingly, algorithms are determining what people are thinking and they have an intrinsic tilt or, or bias because as this digital Frankenstein is saying, here's the next video you should watch, or Facebook is saying, here's the next article you should read. It has an intrinsic bias towards what's the most sensational or radicalizing thing? The center for Humane Technology. Our group is a group of ex technology insiders who worked at these companies. I was a design epicyst at Google, and one of our members, Guillaume Chaslet, is a ex YouTube engineer who helped build the recommendation system. And what he found is that the recommendations intrinsically skew towards something more radicalizing. So if you start a teen girl at a dieting video on YouTube, the autoplay will be anorexia videos. Not because anyone at YouTube wanted that to happen, but because that was better at keeping people's attention. If you start someone at a 9 11, September 11 news video, all of the autoplay results are 911 conspiracies. If you start someone at a NASA moon landing, the autoplay is the UFO conspiracies in flat Earth. And so at scales of 1.9 billion people, YouTube is tilting the world towards what's most radicalizing. And no one at YouTube even knows the languages and cultures of the places they're impacting, including, by the way, at Facebook. If you join a Facebook group as a new mom, one of the really empowering uses of Facebook is you join a group for new mothers to trade advice with each other. And then earlier this year, Mark Zuckerberg said, you know, we need to change Facebook's mission statement. This is actually last year from making the world more open and connected to bringing the world closer together. And so he told his product managers, I want you to help recommend groups for people to join. I want to maximize the number of groups that people join. And so now we have an algorithm trying to predict if you are a mom and you join this group, what's the perfect group I can recommend to you? And of course, what's a really engaging group for moms? The anti vaccine conspiracy theory group. So mothers join that Facebook group for moms, and then it recommends, here's the anti vaxx group. And if you join that, what are other engaging groups from there? You've got flat earth, chemtrails, conspiracies. Right? So we're sort of tilting the world and it has real world consequences because in places like Burma, how many engineers at Facebook or YouTube actually speak Burmese or Sri Lankan? And in Burma, where there's a genocide against the Rohingya Muslim minority group because of fake news being propagated on Facebook, this has real world consequences. If you thought 911 conspiracy theories are bad in English, try taking a tour of 911 conspiracies in Arabic or any other market. So there's no one home guiding the puppet strings except for these Frankensteins, and it's tilting the world in these directions. The second is that technology is getting increasingly better at knowing us better than we know ourselves. This is Socrates who said, know thyself. The YouTube example is one case. It knows that that video is going to work on keeping you there better than you know yourself. Increasingly, I don't even need to know. I can point a computer at your face and know your heart rate and stress level just by pointing a video camera and amplifying the pulse dilation to your blood in your Face. You can pick up people's stress level by pointing a camera at people's cognitive strain from their eye dilation. You can actually figure out how cognitively intensely they're working. If you thought Cambridge Analytica was bad, where you could basically steal personal data to figure out what would motivate people's political voting. Right. Cambridge Analytica was about understanding people's personal political psychology. There's a new paper out from Gloria Mark, UC Irvine, that just by tracking people's click patterns, you can predict the same things with 80% accuracy that you could get from Cambridge Analytica. In other words, I can know your political personality just by knowing what you click on. So technology knows increasingly more about us than we know ourselves, and it's used towards racing to manipulate our attention. And then increasingly, you can use it to manipulate the truth, because technology knows how to fake out the limits of your human instincts. Which of these four video frames of Obama are real? None of them. They're all generated by something called deep fakes. And so increasingly, we're overwhelming the limits of the human mind body system. If you saw what those kids were experiencing at the beginning, there's a limit to what it means to be inside of a mind body meat suit. Same way a mouse has a limited intelligence and can only see so many moves ahead on the maze, there's a limit to how many moves ahead we can see in our lives. And what happens when we're distracted and what happens when we dose ourselves with social validation from friends every 15 minutes? And so what are we going to do about this? I have an idea. Let's go grayscale. Let's just grayscale our phones. Maybe that'll solve the problem. I'm joking because I was one of the people who helped popularize this idea. If you've heard of it, you can actually turn your phone grayscale, which basically reduces some of the addictive color rewards. But obviously what we're talking about here is so much deeper than that. It would be like looking at climate change, where the whole world is heating up and saying, oh, I know, let's ban plastic straws. We need to think more systemically than that. Right? We're going to need to change the system. We're going to need to think more deeply about what's really going on here. And the metaphor I want to give you to introduce what that might look like is I kind of think we're at a period of human history is very much like the beginning of the space race, when we went to space we were so excited about being able to get those rockets and those cameras out into space. We're taking pictures of the outer space world, right. And it actually took us 10 years before we turned the camera back at ourselves and took a photo of Earth. We were so excited. Just like we're so obsessed with technology, we want to build the latest thing out there, we don't want to look at ourselves. Right. So this photo in 1966 was the first. Yeah, the first photo of Earth that that basically the common people saw. And we later took this famous blue marble photo. It helped us see the Earth for the first time. And this showed us the Earth's limits, that we were living on a finite planet with a finite ecosystem. It's hard to imagine now what it might have been like before we saw that photo. Earth is just a concept. We had to see it. And I think with technology, this is what we're doing now. E.O. wilson said, the problem with humanity today is we have Paleolithic instincts or emotions. We have medieval institutions and we have godlike powers. So humane technology. What I want to talk to you about today, how we fix this problem, is going to start by, I don't know if you can see the gray there, looking at an honest appraisal of human nature, that the next phase of our evolution is doing the uncomfortable thing and looking back at ourselves and seeing that yes, we're vulnerable to social validation, yes, we're vulnerable to magician's tricks, and yes, we're vulnerable to supercomputers. And yes, we're vulnerable to algorithms that split test 66,000 variations of toxicity or hate speech to figure out what will animate our instincts. And the way we're going to fix this problem is having this more honest view of human nature, that instead of drilling attention out of people's brains to suck it out and going in this race to the bottom and extracting that there's actually something to protect about how it works. So we want to realign technology with a 21st understanding of human nature. Instead of saying everything that happens inside of a mind and the authority of human feelings. The customer is always right. The voter knows best if those things are up for manipulation. It's as if you debased the source of a authority in a democratic society. So we're going to have to protect it. And that's going to mean a few things. Here's some examples. We're going to have to protect the way that kids develop their identities. Right now, there's these photos on Snapchat. There's these filters on Snapchat called beautification filters because you win in the attention economy. If you're better at giving kids a way to beautify their self image then if you don't. Two photo sharing apps, one offers you beautification filters. They're going to win and out compete against the one that doesn't offer that feature. But we're going to have to say, you know what, we don't want to do this to children. If you see the quote there, she said this is the cutest I've ever been and ever will be. So we've got to stop screwing with kids identity development. We got to stop screwing with streaks. If you're a parent and you know how this works, kids are bought, duped into basically keeping up a streak of 154 days in a row with their friends. So we have to also protect against the mass manipulation of truth and realize that Basically World War 3 is happening in information space and many state and non state actors are using this medium to promote propaganda, conspiracy theories and change the entire belief system of a whole culture. We can do this also by understanding what does it mean to empower ourselves. So what's different about technology? That's the kind we like to the technology that we don't. So back in the 1980s when Steve Jobs was talking about what a Macintosh was about, he said it was a bicycle for our mind. If you think about human being just by themselves walking in the world, their locomotive capacity is limited. But you give them a bicycle and using their same human instincts, they're able to go whole new places and it keeps them involved. They have to use their own muscles, they have to use their own mind to navigate where they want to go. Right? So that kind of technology was good. No one ever freaked out that bicycles were causing conspiracy theories, the rise of populism, paranoia. So we have to go back to understanding what is empowerment and humane technology is possible. If you're looking at how do we leverage human strengths, where has evolution given us an incredible set of Strengths for empathy? FaceTime is a great example of a humane technology because we're built for empathy through our voices, through eye contact. This is great. There's nothing wrong with this. FaceTime is not again pulling apart the fabric of society. FaceTime, excuse me, is not pulling apart the fabric of society. We also have to get better understanding what people are really after. You know, instead of YouTube just trying to maximize what keeps you on the screen and having this intrinsic bias towards conspiracies. What if we asked you, what are you actually here for when you're watching this ukulele video? And if you said, I want to learn ukulele, the computer would be like, great, let me think about what would be the best way to help you. And it would say, here's two friends you didn't know about who play the ukulele who might be able to help you out. It's understanding, for example, that we can easily misunderstand each other in digital communication and text communication. So if someone says, wait, I don't understand. We really need to talk about this. And then the big menu provided to you is, which key do you want to type to get started writing this controversial message? Where it's been shown people are bound to misinterpret each other. And instead, if we say, where is human evolution really strong? Where are we really good at having difficult conversations? We're really good at doing that in person. And you could have new options on the menu that basically let us say, we want, you know, this is where we'd like to talk in person. Or they're nearby right now and you can go talk to them. You can say, for example, that we're vulnerable to something called learned helplessness. There's a feature of psychology where if you show someone, if you take a rat and you put it in a cage and you buzz a buzzer, and right before you shock the cage, the rat learns it needs to jump to try and avoid the shock. But after, there's nothing they can do, they just sit there while the buzzer goes and it just shocks them. It's called learned helplessness. And humans work the same way. If you give us a stimulus, like on Facebook, when you say, hey, I want to tell the world about climate change and I want to share and be so active, what you're really doing is Facebook's creating exponential learned helplessness. Because you have basically people experiencing big global problems that they can't do anything about. But that doesn't mean that we shouldn't do better. We just have to understand that we're vulnerable to that. And instead say, imagine if you did find and replace on those 2 billion people experiencing lone helplessness and said, actually, here's the great work being done on plastic in the oceans, and here's my friend who's working on a thing that's planting a trillion trees using drones. These are real projects out there. But you don't know about that, because technology is just not honoring a sensitive and vulnerable view of human nature. So how are we actually going to transition to this, this might seem impossible. This goes against the business model of maximizing attention, right? Because these business models are whatever keeps people on the screen. And we have stock prices buoyed up by that, how would we ever change that? So we go about changing this with four different levers. So one is that no one actually wants this. It's very simple, right? Once you show people that this is how technology is driving the world, this race to the bottom of the brain stem to pull attention out of people, is that if you show everybody, including the employees, no one wants them that future, no one is excited about that. So it's a collective mission to move from extractive, dirty, burning sort of attentional technology that offloads the externalities onto the balance sheet of society to this greener version. So if you activate the public, and especially parents, by activating the employees at tech companies who don't want this to happen, we've seen there's been enormous changes when employees at the company say, we don't want this for our own children and we don't want this for our democracy. If you saw just Nick mentioned just yesterday, the founders of Instagram, my friends from college, who I haven't talked to about it yet, just resigned from Facebook. And I think that this is really a result of the fact that people don't want this system. They don't want this to be the way. And you can also activate pressure from governments. So when there's massive hearings, that actually helps create the system. And you have to show the fourth thing, that there are alternative, humane ways to do it, that if you start with a vulnerable and compassionate view of human nature, you can fix it. And now I want to show you why this has actually been working. This is not just hypothetical. We've been working on this for a long time and there's a lot of people in this conversation. New survey just came out. 72% of teens think that they're being manipulated by technology. This just came out. A few weeks ago, Common Sense Media did a survey. 13 to 17 year olds believe they're being manipulated by technology. It's much like the cigarettes, where it wasn't about telling people, this is bad for you. It was about recognizing that they knew it was manipulative and addictive and they didn't do anything about it. Second, we've been incubating this concept called time well spent, which the Verge wrote about saying the time well spent debate is over. Time well spent won. Why did they say that? Because Mark Zuckerberg earlier this year said, our big focus for 2018 is making sure that the time we all spend on Facebook is time well spent. Our ideas have actually infected the very minds of the people that are running the industry. So Facebook has made moves, small baby step moves in this direction. Just a few days ago, last week in fact, Apple just announced screen time limiting features with iOS twice so people can actually limit some of their screen time. Google well being just launched a digital well being initiative to help people limit their screen time. These are the tiniest, tiniest baby steps in the right direction. They don't solve conspiracy theories, they don't solve exponential learned helplessness, they don't solve the whole distraction problem. But now companies are saying it's our responsibility and it's now a race to the top for who can care more about the fabric of society. And of course, if you said a year ago to me, one year from now Mark Zuckerberg's going to be testifying before Congress, I would have never believed you. So this change is happening through awareness alone. When you go out there and you talk about this problem and you say this is not the world we want and there's a different way to do it, it actually matters and this is the world we can create with humane. For this world to happen, we do need everyone's help. It will take all of us. It'll take parents to message their friends who are inside of the tech companies. It'll take kids talking to their parents about these issues. It'll take policymakers, it'll take the media. But everyone can be a part of this transition and we need your help so that we can stop exploiting human nature and put humanity's best interests first. Thank you.