How Technology Tests Our Trust

AI is transforming the world, and it.

Starts with the right compute.

ARM is the AI compute platform trusted by global leaders. Proudly NASDAQ listed, Built for the future. Visit arm.com discover. Welcome to the HBR IdeaCast from Harvard Business Review. Hi, I'm Sarah Green Carmichael. There we were, a group of friends driving through Sicily in search of a hilltop village. Our GPS guided us to turn left, then right, then left again. And the roads were getting smaller and smaller. Then suddenly the hillside fell away steeply. There was just a track carved out by the hooves of grazing goats. The GPS said it was a road and we had trusted the technology. Nowadays, though, GPS isn't the only technology people are figuring out how to trust. For instance, there's also a fast growing market of smart speakers like the Amazon Echo and its artificial intelligence Assistant, Alexa Play. HBR IdeaCast Getting the latest episode of HBR IdeaCast, here it is from TuneIn. So far, about 7% of households in the US have smart speakers, according to an NPR Edison Research survey. But of those who do have them, 42% say they're essential to their everyday lives. And those who don't often say they're worried about privacy from hackers or even that the government might be listening in. Devices are becoming so helpful it can be scary. Our guest today studies that tension between utility and fear, and she joins me now to talk about how new technology is testing our trust. Rachel Botsman teaches at the University of Oxford's Said Business School. Her new book is who can youn How Technology Brought Us Together and why It Might Drive Us Apart. Rachel, thank you for talking with us today.

It's good to be here.

So I wanted to just start by asking you a little bit about an example that I think a lot of us have dealt with, which is Alexa. I know that my husband was given an Alexa. He never installed it. It sat on our kitchen counter in the box for three or four months while he sort of was not sure what to do. And in the end he just thought, I don't need this little robot listening to us.

Did he talk about why he didn't want it listening to you?

He was worried that even if it was off, quote unquote, it would never really truly be off. And not listening.

Yeah, it's interesting. Well, he's an anomaly.

Okay, I'll tell him that.

I think he's a Luddite. I mean, I have a husband who's very similar. You know, there's post its over the computer and so he's very conscious of this. But I would say most people, whether it's the device or the app, they give their trust very easily to the device without thinking about sort of the lopsided trade offs that they're signing up for. And this is because technology, what it does is it actually accelerates trust. It makes things very seamless where we don't think about what we're really placing our trust in. But I think your husband's observation was really interesting because how do you trust the intentions of something like Alexa? As a parent, this is something I think about and I think about the responsibilities of companies. So I did an experiment with my daughter where I did plug in Alexa and It took her 48 hours to figure out she could order things. She's three and a half years old, right? Like, so when she realized she has the power of that voice. But by day three, she was outsourcing her decision making. So she was asking Alexa what to do. And then the real one that got me was, what should she wear? Because no one interferes with what Gracie wears. Like, definitely not. And what she didn't realize was our version of Alexa also had a camera.

Aha.

And Amazon have introduced a very helpful style check feature that if you stand in front of it in two outfits, it gives you a rating, but it also makes suggestions of the cardigan doesn't go. Or maybe you'd like these shoes. Would you like them shipped to your office or your home or your school in the next hour? This is the point. It's like you understand that from a commercial point of view, you can see the company going like, this is the way for us to get into fashion. But should these companies have to be more upfront about their intentions that it's not a helpful female assistant? There is a commercial agenda.

Well, and why are so many of these assistants female?

It's funny you say that, because when I introduced Alexa to Grace, she said, is she like Siri? And then she knew Viv as well and she said, they're ladies, not women. I mean, that's how ladies. I thought, man, like, she's aware of this. But it's interesting from a design perspective. So you see this with the self driving cars, right? They're Iris, they're female names. And it's really interesting when you look at the gender studies that just by changing the name or making it feel more female in the design, people are more trusting towards that technology and more forgiving. So MIT did a study where they changed the gender of a driverless car and when the car did something wrong, they're more forgiving when it was female versus when it was genderless or masculine.

You've said that technology accelerates trust. I think that many people would expect it to be the opposite. Why do you see technology actually accelerating the formation of trust?

When trust forms between two people, deep trust, meaningful trust, enduring trust, it actually needs a little bit of friction. That's human connection it needs. Trust is really built when things go wrong. Like when you work through those things and you go through that conflict resolution. You see this with teams, right? When they've gone through some crisis or a product they were working on fell, and they came out the other side, they have higher levels of trust. But what technology does is it accelerates that process to the point that it's like trust on speed. And so often we click, we accept, we share. We make a decision as to whether something or someone or a piece of information is trustworthy based on very limited information. And so efficiency is kind of the enemy of trust. And so technology doesn't like friction. But friction is often human connection. Friction is when you consciously slow down and start to ask whether this thing is worthy of your trust. This isn't new because we've seen this play out with information, like the way we share information. You know that wonderful study that was done that 80% of people shared a piece of content just based on the headline? That's a really good example of just giving our trust away too easily.

You say that people have two ways of sort of arriving at trust. One is through emotions and one is through reasoning.

Yeah.

Is one of those sort of better than the other, stronger than the other?

Yeah, I mean, it's the classic head and heart thing or rational and emotional. And I think it's really interesting because people often say, you know, when it comes to trust, I rely on my gut. But there's certain situations where we don't ask enough rational questions. And so there's a science behind what makes someone trustworthy. And there's four traits to it. So there's competence and there's reliability. And that's really the how side of the equation, how well you're going to do something. And then there's the why side of the equation, which is integrity and benevolence. So how much do you care? And I think the integrity is the most important piece, which is, do your intentions align with mine? What are your motives? And the interesting thing is that we think of interviewing an employee. A lot of questions are often on that how side of the equation. And then we kind of rely on our Intuition on the why side of the equation. But when I think of when trust is broken down in my professional and personal life, it's usually because there's a misalignment of intentions. So I think we need a more rational process for actually asking questions around the why side of the equation. And that, I think, is more balanced in terms of combining the emotional and the rational in decisions and who to trust.

And how do you see technology either helping or hindering that kind of trust development?

So it's interesting. So one of the questions I ask myself is, is technology making us smart about who to trust, or is it encouraging us to place our trust in the wrong people in the wrong places? And I think it's a little bit of both. There's a great trust theorist called Diego Gambetta, and he talks about the two enemies of trust, one being bad character and the second being poor information. Now, you could say, well, technology is going to not make bad people good, but it can make people more accountable. So we see this with ratings and reviews, but it's the information bit where the power really lies. Like, how can we give people a more complete picture of who this person is? Where I think the technology needs to stop is it shouldn't make the decision for us, so it should provide us with the information. But that you want enough friction that you actually think about that person, that you're hiring that person, that you're recommending, whatever it is that that's still a human decision.

You have an amazing story in the book of someone who took care of you when you were a child, who was not at all what she purported to be.

No.

First, I would just love to hear you tell that story, and then maybe we can talk about our judgment calls really worth leaning on at all anyway, if humans can be so flawed.

So what happened was my mom wanted to go back to work, so she hired a nanny for us through this posh magazine called the lady, which is bizarre in itself. This woman called Doris got the job. And mom said that one of the reasons why was she had this very friendly Scottish accent. And she said that she was a member of the Salvation Army. And I remember her walking in the door, and she had the uniform and she had these glasses, and every time she said my name, she rolled her Rs. And now these things, they are trust signals. So we all use them. These are clues or symbols that knowingly or unknowingly we use to make a decision as to whether you're trustworthy. And the problem is that some signals are Louder than others or we tune into the wrong signals. So she tunes into Salvation army and Scottish accent. Now, what was amazing was that she lived with us for 10 months and she was a good nanny. But then she disappeared and my dad started to worry on the third day when she didn't come home. So he went round to our neighbors because their nanny and our nanny were friendly. And the neighbors said, oh, it's so funny you've come round because I've just found out your nanny and our nanny are running this drugs ring in London and they don't belong to Salvation Army. And then a few days later, the police came around and knocked on the door and they arrested my dad. Yeah, because she'd robbed a bank, she'd been in an armed robbery, and she was a getaway driver, but she'd use our silver family Volvo estate.

Oh, my goodness.

And so now my parents are like, well, I hope this woman never comes back. But what's so interesting about that going, she was competent, she was reliable, you could even argue she cared. But her intentions were not aligned with my parents.

Definitely not aligned, no.

So what was really interesting to me was like, how did they make such a bad decision? And what they realized is that they faced a trust gap. They thought they had enough information to make a decision about her when there was a lot they clearly didn't know. And so I started interviewing the babysitting platform. So urban sitters, Just one. And they're remarkable when you think about it, because they are, you know, okay. Renting out your homes is amazing. Finding dates is amazing. But finding a stranger on the Internet to look after your children is really, I think, an amazing risk frontier to overcome. And what I discovered is that 75% of all applicants who apply never make it to the platform because they're using machine learning and all kinds of data intelligence to, you know, did you really get a child care diploma? Do you use absolutely foul language? Do you have a criminal record? All those reviews and ratings and referrals real. So the bar is really high. And so I said, would Doris make it onto platform? She was like, no, no. Your parents would have known that she wasn't a member of the Salvation Army. And so this, I think, is a good example of information that, if it's presented in the right way, can help us make better decisions about who to trust and dodgy nannies to avoid.

Is there another company out there that you feel like is doing a really model job of maintaining trust?

So the number one question I'm asked by companies, how do we rebuild more trust? How do we rebuild trust? Or how do we gain more trust? And we, with all due respect, I think it's a terrible question. What you want to be saying is how do we become more trustworthy? And you want to really understand where trust lies in your organization. It's across your organization. Trust is really part of your DNA. So I think Tesla actually do it really well. And the reason why they do it well is they're telling you, continually telling you a long narrative around where they're going. And it's deeper than purpose. Right. You know the why behind that company, but you can visualize them taking us to Mars and you can visualize the self driving cars. And then when something goes wrong, like the accident with the self driving car, they respond really quickly. They get into the details of what happened in the accident, they own that mistake and then they communicate what they've learned from it. And so you don't feel like it was just the car went wrong and they've moved on. You feel like things are changing and evolving and it was towards sort of a greater goal of human safety and self driving cars. So I think they are really good at thinking about trusting their brand and understanding what impacts it. Beyond Elon Musk and the brand itself, but the actual behavior of the company.

One of the tensions here that companies really have to figure out how to manage, I think is on the one hand, they need to authenticate that people are who they say they are. But then the primary way of doing that, it seems like, is for people to give up a ton of personal information. And then it's almost like the stakes spiral up even higher because now the company owns all this data about you that they have to keep safe. So it feels in some ways like trying to become more trustworthy has also made companies at greater risk if there is a data breach. How do you think about all of that?

It's a really great question. So I talk about in the book a trust hierarchy of needs. Identity is the base level of the stack that you have to know that this person is who they say they are. But the result of that is, and this was the part that really amazed me was how much data there just is on us. It is surveillance. I mean, there is no other word. And I think what struck me about the Equifax data breach was the number of people who didn't even realize their data was being stored. And so I understand how we've got into this conversation around transparency being the end goal. But when you think about it transparency and trust are not synonymous, because if we need things to be transparent, we've kind of given up on trust. Think about teams, right? Low trust teams is when the boss needs to know everything that's going on. And I think we're putting the same demands on companies, and I wonder if this is the right end goal. And so I think, is it more about visibility? Is it more about control? Is it more about that the company is extracting a lot of value from this and we can't see what that value is. So I'm not sure transparency should be the end goal of this. Hmm.

Ethan Bernstein at HBS did some research on how increasing transparency sometimes makes people feel less trusted.

Definitely. And it's an important. It's almost an impossible goal for companies to maintain. That's where I was going with this, that you set the bar so high, because then if you keep anything hidden and you say, I'm a transparent organization, that is the enemy of trust. So it's like when people say, we're completely secure, it's like, no, don't promise that. Right. Because you can't make that promise. So there is a direct correlation.

Some people have sort of held up blockchain as this sort of technological fix that will finally fix this and make all of our data more secure.

Yeah.

Is it really that?

I don't know is the answer to your question. I mean, the thing that excites me most about the blockchain is less the data piece, but is more around the transfer of value. So if you think about how hard it is for you and I to exchange an asset, and this relates to the data. Right. There's lots of intermediaries in between, and those intermediaries are the ones that then extract the data from that exchange. So I do think in that sense that probably within the next five years, we'll see a transformation in asset transfer. And then, you know, you look at countries like Estonia and even places like Saudi Arabia, Israel, and you're starting to see the early signs of government, that when you're born, you'll have an identity that sits on the blockchain and you'll have your own private data locker, and you'll be very much in control in that you will be able to give permission to companies and institutions to access that data. And I think that's coming, but I think that's a longer time frame around this.

You have a really great example in the book about China and what the sort of Chinese state is doing in citizen surveillance.

I started tracking this around 2014 because the Chinese government, they issued this policy that basically said that by 2020, every Chinese citizen would have an. It was called a social citizen score, but it's essentially a trust score. And it was really interesting tracking this. There's an amazing professor called Roger Crimea who translates all the documents. Basically, they came out with the economic argument that people in China don't have credit scores. There's high instance of fraud. And now what we're seeing is that they're tracking online purchases, they're tracking what we say online. If you say something negative, my score goes down. And they've started to introduce the penalties to this. So you can't apply for certain jobs if your trust score goes below a certain level. Your kids can't go to certain schools. They just ban more than 6 million Chinese from taking airplanes.

Rachel, this has been really fascinating and thought provoking, occasionally terrifying, but also hopeful. So thank you, Rachel, for being here today.

Thank you for having me.

That's Rachel Botsman. She's the author of who can you trust Now. If you are wondering whether to trust Amazon's Alexa, try this. Build a passionate company. To build a great business, companies need a purpose. One that transcends the traditional bottom line. It's HBR's management tip of the day and you can access a new tip every weekday for free if you have an Amazon Echo. Thanks for listening to the HBR IdeaCast. I'm Sarah Green Carmichael.