The Economist: Does AI mean the end of jobs as we know them?

Hello, my name is Sebastian Thrun and I have the great pleasure and privilege to be with Zayd Enam, the CEO of Cresta. Hi, Zayd 

Sebastian: Hello, my name is Sebastian Thrun and I have the great pleasure and privilege to be with Zayd Enam, the CEO of Cresta. Hi, Zayd 

 Zayd: Hi Sebastian. 

Sebastian: Cresta is a company that brings AI into corporations for things like sales chats and customer services. Zayd, tell us what you’re doing? 

Zayd: Yeah, absolutely. We’ve built a platform that learns from the very best of every conversation and is able to then coach people in real-time, both the agents and the managers to ultimately deliver the most effective conversation to each customer. Our goal is to ultimately help every representative of a company to be the very best representative by coaching them in real-time, based on what is the most effective conversation to have to ultimately drive the best business outcome.

Sebastian: And Cresta is an artificial intelligence company, right? This is artificial intelligence

Zayd: Yeah, absolutely. As you know, we started at the Stanford Artificial Intelligence Lab project and we’ve built some of the leading NLP and artificial intelligence technology to power the platform.

Sebastian: When people hear of artificial intelligence, they think job replacement, they think you want to replace people. When you learn how to converse with a customer through Cresta, why are you not replacing people? 

Zayd: Yeah, that’s a really, really great question. I think that’s a common view and maybe perhaps a lazy view of artificial intelligence.

I would say that. There’s a very lazy view that artificial intelligence basically can do anything a human can do, but automated. But I think the very exciting applications of artificial intelligence are combining both humans and machines in new creative ways to unlock capabilities that we’ve never even imagined before.

So, if you look at say the advances that we made in terms of phones and computers, they haven’t been sort of figuring out what we’ve done as humans and automating them. We’ve figured out how to use those fundamental technologies as building blocks to unlock and enable entirely new applications and new sorts of perspectives that weren’t even possible before.

And so that’s really what we’re hoping and what our perspective is with artificial intelligence. Is not really going for the lazy way, which is figuring out what can be automated of an existing process, but how can we create new processes and new combinations of humans and machines that are entirely able to build upon these building blocks and create entirely new value and new sort of advantages for us, as humans and as companies.

Sebastian: That’s exciting Zayd, walk me through your customer journey, who is your customer? How do you get on board? What do you do for those customers? What’s the response of the people involved? 

Zayd: Yeah, absolutely. So, we really took an early goal to focus on serving the enterprise. And the reason is because we found that for large enterprises, there’s a very big sort of operation customer-facing operation often in the contact center, and that’s really where we focused initially because we found that there are so many pains and opportunities in the contact center. And so what we’ve done is we both have the technology and enterprise-grade platform to serve the enterprise.

What we are able to do is identify a specific use case, so that use case might be a sales use case, or a care use case, or a retention use case where the contact center is having conversations with their customer and are looking to sort of ultimately either drive a sales conversion, resolve a customer issue with better customer satisfaction in a very efficient way, or ultimately prevent a customer from canceling or churning. And so what we’re able to do is build a model, build an artificial intelligence on a platform that can then learn from the conversations of the very best representatives for that use case, and then deploy that so that every agent, every representative is getting real time coaching real-time assistance, live on a conversation to help them perform and act as the very best.

And so the results have been tremendous. We serve fortune 500 enterprises like Intuit, Cox communications, Porsche, and large, large enterprises that have these needs. What we’ve been able to do is find these specific use cases, land, and then the most interesting and most inspiring thing for me personally, is that the businesses not only find value in terms of business outcomes but the people that use the software, they actually love it.

Not only do they love it, but they also love their career even more for it. So they’re using the application, and the results, we do a study and we do a survey at the end of each deployment, where we ask how does Cresta impact your love for your career? And the data that we get is that 90% of customers who use Cresta their, their love for their career actually increases. It actually increases for two reasons. One is that they now have the technology to basically help automate the menial, tedious tasks that we’re usually doing. So automating the basic form, filling applications, or automating these kinds of things that, that they don’t really want to be spending their time doing a very boring part of their job, but then secondly, they also now have the confidence to answer customer inquiries and have a customer conversation by getting that knowledge that makes them feel that they’re ready and primed to be successful in the role. So the company is really investing in them and helping to give them the tools to be successful in the role.

When you feel that kind of support and that kind of confidence in being able to interact with a customer that ultimately makes you happier in your job and makes you primed to succeed. That’s really the impact that we’ve been seeing, which is personally very inspiring for me.

Sebastian: That’s interesting because normally AI is associated with job replacement, but you are telling us you’re leveraging the AI to make your customer’s agents stronger and making them love the job more and stay in the job longer.

Is that correct? 

Zayd: Yeah, absolutely. Absolutely. As we mentioned you see this impact on retention of the representatives as well. In this industry, it’s very common for people to leave a job after a year, their churn in call centers is about 33%, that’s the industry average churn, but what we found already, and this is anecdotal, we’re doing these studies right now, is that we find when people are more successful when they have the tools to be successful they stay at the job longer. So agents are staying at their job for longer than a year. You’re staying for multiple years because the company is really invested in your success.

Sebastian: I want to change gears a little bit and ask the questions. What’s in it for the world? Like, what are you learning about artificial intelligence that surprises you and what should we think about AI as a result?

Zayd: Yeah, that’s a fascinating question. I think a lot of sort of things that we’ve found or we’ve learned about artificial intelligence and its applications in the world over the last 10 years have been very surprising. I think one of the biggest ones is that often it’s really hard to build in these kinds of really high-risk situations to build fully autonomous, highly accurate artificial intelligence to really serve the need in a fully autonomous way.

That’s really where augmentation is sort of the right mix of humans and machines can have a really big impact and make things possible. We’ve seen this in terms of tools that help sort of doctors, drivers in our case, agent representatives, be more effective because the technology is artificial intelligence.

While it is made significant progress in terms of all the work in NLP and computer vision, we’re still not at the level of human-level intelligence or general sort of reasoning. And so that’s really why we need to be able to combine that technology with humans and really make the best of both to sort of solving these really high-impact use cases, across many different industries.

Sebastian: There’s a lot of people that warn of artificial intelligence as enslaving us. Like you don’t, Elon Musk says they’re going to eat us like chicken sandwiches. Stephen Hawkins talked about this, Bill Gates, and so many others. Are those people, right? Is it a matter of time until you get there? Or do you think they’re wrong?

Zayd: That’s a great question. I think there’s a lot of sort of media sort of a frenzy around that kind of direction. I think Elon Musk and sort of folks like that are very intelligent. There’s a lot of very intelligent folks that have had very sort of grandiose statements about sort of the enslavement of what artificial intelligence is going to be.

I think fundamentally artificial intelligence is a technology just like any other. So just like how the phone has made it possible for us to be superhuman or a computer has made it possible to, for us be superhuman in many ways and be able to use that technology to help us do better. Artificial intelligence is, is just much like that.

It’s technology, that’s still, in some cases in its infancy, it’s already shown dramatic impact, but there’s still a lot of potential for it. It will be a technology that we’ll learn to as humans shape and build to our needs to ultimately serve us effectively. I think the hyperbole around it is sort of a disservice in a few ways. One is because I think it causes a lot of fear in the public and in terms of governments and all these things, which I think that fear right now is not currently well reasoned and I think there’s an opportunity to build artificial intelligence in a safe manner, in a manner that serves humans.

And I think that’s really what we want to focus on is that how do we build this technology? Just like we built the computer and the phone to serve us and that helps make us more effective and make us more productive. I think that’s really the opportunity for us to focus on.

I think there is a big impact in the future, whereas we’re maybe to discuss is like, how does artificial intelligence impact economic inequality or how does it impact sort of war and manufacturing, of goods in war. But I think the sort of hyperbole around sort of enslaving us is a little bit sort of going far ahead.

Sebastian: Let’s talk about this a little bit more. One of the doomsday scenarios that comes creeping in here is this idea that AI can take over your, the repetitive part of your work, the stuff that you do in and out.

Like, if you’re an accountant or if you’re a taxi driver, and you kind of do the same thing in and out, even medical doctor and diagnostics. What should people think that go to work and say, I pride myself on being, reading me to good, and doing the same thing over and over and over again?

Zayd: Yeah. I think that’s, I think that’s going to change. I think there is I think in general, a sort of a big opportunity for, for a shift in the way people view their value and their sort of impact on the world. In the sense that I think we all need to adopt to have a very strong learning mindset. I think the world is going to dramatically change and it’s only accelerating because of the rate of technological progress. If you look at the technological progress in the last a hundred years versus the last thousand years, we’ve been exponentially accelerating in terms of what’s been possible with technology. So in an environment like that, it’s dramatically different in terms of what the expectations are.

I think just the mindset that we need as humans and as people is a very different mindset. We need a learning mindset that always recognizes that the world is going to be different in five years, it’s going to be different in 10 years, and we want to have that ability to grow, to upscale ourselves, to learn new things, and learn new skills to sort of keep up with that.

And it’s just a very different value system, I think than what was, what we had a few hundred years ago. 

 Sebastian: That you are one of the leading technologists in artificial intelligence. You’re running a thriving company that’s getting great business traction. But stepping a bit aside what’s your advice for a CEO is like, if you are a CEO or a business leader, a C-suite executive listening to this what would you take away from me, but for you to do?

 Zayd: I think at the end of the day, everything in business, everything starts from the customer. It always comes back to what can you deliver for the customer?

What is the experience? What is the desired business outcome? What can you do for your customer? And then walking back from there in terms of what is the product, what is the technology that you want to sort of leverage to be able to deliver something better for your customer, a better product, a better experience, a better process.

I think that’s always where we need to start and the way we need to think about it. So I think if we do that and we view this artificial intelligence under that lens, and what we want to do is go back and look at every part of our product experiences, whether our product, a car, or software, or if it’s a process and consulting service, we wait for one to identify what is the job that the customer is looking for us to help complete.

So as a customer, looking to sort of getting from point A to point B, are they buying software to get to a certain business outcome, or are they consulting with us to get advice. That job, how can we leverage technology to help them get that job done?

I think the really interesting thing about artificial intelligence is it unlocks this massive new set of capabilities that make it easier to close a gap between what is the job the customer is looking to get done and what’s possible with the technology and the product. 

I think if we focus through that lens of what can, what are new capabilities are possible with our product that can make ultimately the job that customer wants to get done? More effective, more efficient, cheaper, better, then I think that’s really the sort of alignment and the vision. I think a CEO that can most effectively deliver to their company and to their customers. I think that’s the place to start. 

 Sebastian: We obviously just in the middle of a horrendous pandemic. And we see unemployment rates at levels, never before.

We see a lot of people in the hospitality industry, massive disruption in transportation and others, but it’s also an accelerant. So if you are a CEO of a company that is not very technology-focused, And you hear about this thing, artificial intelligence. What is a practical way to get involved, to, get on the bandwagon, and be on top of this?

Zayd: Yeah, I, think there’s a really interesting opportunity here, because if you look at certain places, for example, you look at India as an example. India. And some places like Pakistan, the country I’m from, if you look at the last 10 years, they adopted mobile phones at a much, much faster pace than anything that the United States did.

The reason that happened is that they leapfrog the whole sort of wireline, the whole wireline phone. And the whole broadband phone. And they sort of had that there was a fundamental change in wireless technology that those countries were able to adopt that leapfrog that entire sort of wired infrastructure and at a much faster pace, adopt that technology much cheaper service to deliver to the customers.

I think there’s a similar opportunity here. The opportunity is that there’s been a fundamental change in technology that makes a whole new set of capabilities possible. That technology allows even companies that were behind in the last five years to leapfrog their competitors by adopting that technology and realizing how it can change their fundamental process, their fundamental product, and deliver ultimately a better experience for their customers.

I think that’s really the opportunity here is to completely go back, go to stage one, recognize that there’s been a disruptive change in technology, and examine it. How would you re-evaluate if you were to build a company and build a product from the ground up, leveraging the latest technology that you have available to you now, how would you reinvent everything from scratch Then use that as a starting point to then go back and figure out how to take your company from where you are right now to the future. 

Sebastian: That’s a radical statement, but I’ve myself seen it happening. The main thing that comes to mind is Randall Stephenson. When you talk about AT&T  and turned AT&T from a landline into a mobile company, it was just brilliant.

And I think it was exactly what you said is that where someone sat down and said, what should our business in the future be? And let’s just build that business. It’s very, very hard to do.

I have, I have one last question for you is that you’re not on stage as a world-recognized leader in artificial intelligence. You’ve created one of the most thriving startup companies in that space, but it’s not too long ago that you were a student enrolling at Stanford University, trying to figure out what to do with your life.

Now that you live at older tell us what advice you give to young people. People who are entering the world, maybe not even technologists like yourself, but maybe humanities people or social science, people that want to do a good thing for the world in the future. What should they do? What should they know about what advice do you have for them? 

Zayd: Yeah, absolutely. I think there are two pieces of advice Sebastian as my mentor that you’ve had that have had the biggest impact on my life personally. I think the first piece of advice, I think early on in our project you gave me the advice that we were working on this technology. We were sort of figuring out where we’re artificial intelligence has the biggest impact in the world. You told me that basically what you need to do is you need to go to your proverbial desert, because you told me that when you built a self-driving car that you started with, basically going to the real world, figuring out how to build a car that works.

Then once you’ve made that technology made that advancement, then you came back to the lab and figured out what was the advancement that the scientific advancement that was made. And I think that approach of going out into the world and figuring out how to build technology.

So we as scientists and innovators, aren’t just stuck in our academic labs and isolated from the world that we’re actually out in the world and sort of figuring out how to leverage our technology and make it impactful and useful for the world.

And then coming back and figuring out what is the advancement. I think that’s always been a piece of advice. That’s really stuck with me for a very long period of time. And that’s been key.

The second piece, I think for me personally, is that one thing was you’re right in the sense of modern education right now, these days, it teaches people to focus on being smart. Like what is the right answer to that test? What is the right essay to write? What is the math? How to maximize scores and SAT all these things? It teaches people how to get smart, how to get good, but what it doesn’t teach is empathy. And I think the biggest thing, I think the biggest thing that matters a lot in terms of being able to collaborate effectively with teams to lead people, to be able to do things effectively, it has to start with empathy.

It has to start with being able to see not just your own perspective, but the other person’s perspective. I think that’s the same thing for building teams, the same thing for working with customers and partners. I think all that starts with empathy. And I think that’s a big piece of advice is that.

Education right now is entirely focused on sort of being smart, getting the right answer, but less focused on what is the other person’s perspective and how do you communicate in a way that’s not just correct, but in a way that’s effective, in a way that understands that person’s perspective and ultimately gets both of you to a more effective destination.

That’s the second piece of advice I would have is that learning how to do that is really important. And I think the only way to do that is by just exercising and trying it, and recognizing that schools and universities won’t ever teach you it, or won’t ever optimize to sort of have you learn that you need to go figure that out for yourself.

Sebastian: Yeah, well coming, coming full circle in this, in the unit. I love what you say that since you and I are building with Cresta coaches, that coach people to be more effective in a job, does your AI learn empathy? Does your AI help people be more, have higher levels of empathy? 

Zayd: That’s interesting. And that’s something that I’m personally very fascinated about is that when you coach people, can you help develop empathy? Can you help develop that skill of empathy?

I think you can, because I think just like a human coach, a human mentor can help develop empathy in a student or a peer or a sort of someone. Someone following them. I think the same way software can help coach people to develop those skill sets. And I think there’s been a lot of really great work in the eighties and nineties at Stanford, actually, in terms of what is the etiquette of a computer? What do people expect from a computer in terms of their interactions? 

It turns out our expectations from a human perspective is we expect the computer to be able to talk to us just like a human would. And like, that’s ultimately the best experience for us. So if you look at the very best coaches, the very best sort of coaches and leaders, they’re able to develop those skillsets within their people. In a way, that sort of is effective at its sort of teaching them things instead of telling them things.

I think that’s one of the lenses that we really view our coaching and our software under is how do we help coach people to that behavior instead of telling them what to do. And I think that’s an important part of it, but I think that’s something for us to sort of shape in the future. 

Sebastian: Zayd mentioned that I am and was his mentor at Stanford and now at Cresta, and every mentor has a ceiling. One of the most amazing things in the life of any mentor is if your mentees break for your own ceiling and I think that’s what’s happened.

That is an award by a worldwide leader in artificial intelligence. He’s running an amazingly fast-growing company with Cresta, a massive number of new customers just in the last month, and is very thoughtful and helping us understand AI.

I hope you can take from this interview that AI can be your friend. It’s a technology like many other technologies. It can be good for good or bad, but when used for good, it turns people into better people, into stronger people, into better experts, into more competent people, and then people feel that they will emerge and say wow, I love having an AI coach helping me be better.

 That’s it all the press says about it. And I think that’s what they should be about. And the world going forward is that I want to thank you for this conversation. It’s been a great pleasure and looking forward to many, many great years of future collaboration. 

Zayd: Thank you so much, Sebastian. And likewise.