The workless world with Calum Chace – episode 10

Episode 10 summary

December 3, 2024 // 28 min, 05 sec

Experience says we tend to overestimate the short-term impact of a new technology, but underestimate its impact long-term. In all the AI hype of today, are we underestimating the seismic shift it will bring to the workplace of 2050?​​

In this episode:

Marissa is joined by advisor, author and co-founder of the London Futurists Calum Chace to explore the longer-term impact of AI on work and society, and what it means for our sense of purpose and identity.​

• Our job roles have historically been key to our identities, but where will we find meaning in a world without jobs?​
• What will be the last jobs to be automated?​
• How will ‘work’ and ‘leisure’ be redefined in our lives in an AI-powered world?​
• Will this be a world of AI-haves and AI-have-nots? How will we share the value AI creates?

Episode transcript

Marissa Geist 

Welcome to “The talent time machine,” the podcast for talent leaders that takes you on a trip to the world of work in 2050. We're going to think about the trends, possibilities, and new realities for talent. And today, we'll be exploring how AI challenges the sense of identity, purpose, and motivation we've come to rely on from work. I'm Marissa Geist, and I'm joined by advisor, speaker and author on AI and the future of work… Calum Chace. Welcome, Calum. 

Calum Chace 

Hi, Marissa. Very nice to meet you and nice to be on your podcast. 

Marissa Geist 

Before we jump into the content, I was looking at your bio and I noticed you came from consulting and then into the world of future-predicting. So tell me a little bit about your journey. How did you end up out of the big four consulting and into this more future facing world? 

Calum Chace 

So after 30 years in journalism and business, I retired. And turns out I'm not very good at retiring. I don't play golf, so that doesn't help. I, had long been interested in AI and I thought, well, I'll take that up as a hobby and read more about it. And I started writing about it and it got more and more interesting the more I got into it. 

And I read a couple of books and people started asking me to explain AI in talks, in keynote talks, at conferences and things, and I thought, well, that's great. This will be really good marketing for the books. And of course, it was the other way around. The books turns out are good marketing for talks. And I've been doing that for a bit over a decade now and really enjoying it. 

Marissa Geist 

You've been on the AI journey kind of from the beginning then. What's changed since you started? 

Calum Chace 

Oh, an enormous amount. My timing getting into was immaculate and entirely by accident, of course. I started just before the Big Bang, the first Big Bang in AI. AI actually got going just a bit before I was born in 1956. But really, nobody made any money out of it until the first Big Bang in 2012, when Geoff Hinton and some colleagues got what was previously known as artificial neural networks to work effectively. 

They rebranded it as Deep Learning. So Deep Learning took the world by storm, and big tech companies produced loads of really great services, mostly on smartphones, which people used. And they are basically miracles. The fact that Google Maps, for instance, can tell you how long it will take you to get from, say, where I am now in Andalusia up to Madrid to within ten minutes, and it will tell you where to diverge if it's a car crash and so on. 

That is fundamentally a miracle. And then, of course, the other thing that's changed is the arrival of Transformer AIs, which we all know as GPTs. So there’s GPT-4, there is Mistral, there’s Bomber and so on. And that's been another revolution. In fact, that was the second big bang in AI that happened in 2017 and we all noticed it in 2022 and ‘23. 

And that's really shaking the world up again. So it's exciting times. And I'm sure you're finding that in your work as well. And I'm sure it's the thing everybody's talking about. 

Marissa Geist 

It is. It is definitely the word of the year, if not maybe the word of the next couple of years. I did listen to your podcast where you talked about how Neil Armstrong made it to the moon with little more than what a good toaster does today, as far as how the technology is moving and moving at such a fast pace, I just I think it's incredible. 

Calum Chace 

Yeah, it's completely bonkers.  

Marissa Geist 

So I think one of the most interesting things I want to talk about today with you is the difference of work as it changes how we identify ourselves. So when we talk about today, sense of purpose, even social ranking, everything kind of derives, unless you are a baby, an aristocrat or a retiree…  

From what you do to make your living. But you're arguing that the future looks very different. So can you fill us in a little bit on how you think about what the future looks like? 

Calum Chace 

Yeah, I can tell you've done your homework. So actually, I'm not sure I agree with your premise that we all get our meaning and our purpose from our jobs at the moment. And this might be a European-American thing. I don't know, because at least the conventional wisdom is that Americans take jobs more seriously than Europeans. I think Americans have this view that we Europeans just sit around drinking cappuccinos in sidewalk cafes the whole time and work about three hours a day. 

I think even Americans, to be honest, don't get the real meaning, the real purpose in their lives from jobs. I think we get our real meaning from our families, our friends, our hobbies, our interest, our religions, our fundamental beliefs. 

But what jobs do for us is they certainly give us structure in our lives, and they give us something to get out of bed for in the morning. And they give us a thing to call ourselves where we meet each other at parties, the first thing we typically say is, “what do you do for a living?” So they do a lot for us. 

They do a lot of important things for us, and we do need to think about what happens if and when, and I'm sure we'll come on to the reasons for thinking this, if and when machines one day take all the jobs. And so there are no jobs for humans, which I think is going to happen. I think that's a minority view at the moment. 

But I think it is going to happen, and I think we ought to be thinking a lot more about how to prepare for it. 

Marissa Geist 

One of the other podcast speakers I had said thinking about the future, especially the future as far as 2050, the ideas should seem absurd, ridiculous, fantastical. Otherwise, they're not the future. Because the future is not just a continuation of today. So it does sound absurd that there would be no jobs, jobs as we know them today, in the future. 

But in your time frame, are you thinking, you know, we talk about 2050, is that the time frame you're thinking of? 

Calum Chace 

Oh, yeah. I think that a jobless future will arrive before 2050. I used to think that it would be perhaps a little bit after that, but, GPT-4 and the rest of them, the rest of the large language models, have accelerated my time frames along with everybody else's. There was a really big survey done last year. It was the third in a series of surveys of AI researchers. 

In this one I think there are about just under 3000 respondents. Their time frames—and, you know, these are people working at the cutting edge of it—their time frames in some cases have halved. They used to think certain milestones would be hit like, you know, human-level AI say in 2060. And now they're thinking 2040. I mean, things are really telescoping down. 

So it is impossible to tell how long things are going to happen. You have to always bear in mind that we overestimate the impact of any new technology in the short term, and we underestimate it in the long term. There's actually a name for that. It's called Amara's Law, named after Roy Amara. And technologies can only be really deployed once humans have got used to the idea and had a few cups of coffee and thought about it for a bit and worked out how to deploy them. 

So it always takes a bit longer than you think. But you get there in the end. So will we have a jobless future by 2050? I think so probably. Will it happen by 2040? 

It could do. The speed that things are going at at the moment is really quite exhilarating or scary, depending on whether you think it's good or bad. 

Marissa Geist 

Both at the same time. I have four boys and they range in age from 21 to 9. Even in their lifetime, the idea of what you do at work, how you do work has changed drastically. I think about my nine year old, he’ll certainly be in whatever the world of working or gainfully being a human looks like in 2050. 

He wants to know what should I do? What should I do? So maybe work, that narrow definition, because I agree with you, having your total identity come from work is probably minority versus a majority view. But that idea of like, what do I do with my time every day? How do I earn a living? Well, if there are no jobs, what do people do? 

Lounge around all day? Bonbons, soap operas or something else? 

Calum Chace 

So that is the $64 million question. And there's two parts to the question. One is how do we all get access to the resources that we need for a very good standard of living? Because if we all don't have jobs, then under the current model, we all starve, and that's not a great future. And then the other big question is what do we do? 

You know, how do we fend off boredom and how will we find meaning? 

And I actually don't think that's a terribly difficult question to answer. It's important to distinguish between jobs and work. Jobs and what you get paid for and work is things that you do which you might get paid for or you might do it for fun. 

So if you build a matchstick model of Notre Dame Cathedral and it takes you 10, 20 years to do it, but nobody paid you for it, then that's work. But it's not a job. So I think humans will always work. We're wired that way. We've had hundreds of thousands of years of evolutionary training to work to achieve whatever goals we have, and the work might involve reading and writing about AI, for instance, or playing golf or exploring or, you know, having fun. 

But that can be work. But the really difficult question is how we distribute resources to everybody so that everybody has a great standard living in a new world in which there are no jobs and presumably either governments own everything or rich people own everything. All the assets, including the AIs, which are doing all the jobs to make all the goods and services for the world. 

And it is absolutely not immediately obvious how we do that. I'm a capitalist. I don't want to see the end of capitalism because it's achieved incredible things. The market is a brilliant mechanism for allocating scarce resources, so I think the future should probably still be capitalist. But how you get the resources not necessarily entirely evenly spread, but spread so that everybody has a great time? 

It's not immediately obvious. I’ve named the future I'd like to see “fully automated luxury capitalism,” in which we have an economy of abundance and everything is really, really cheap. And so it's easy for resources to be allocated differently. But that's a minority view. 

How do we get from here to there? Very tricky. And we need more people thinking about it. Not enough people are thinking about it right now. 

Marissa Geist 

That assumes that we fall into a fully automated luxury economy and not into a fully automated luxury dictatorship. 

Calum Chace 

Or dictatorship, or just terrible, just terrible dystopia in which a tiny number of people own everything and everybody will starve to death, which is, you know, an outcome devoutly not to be desired. Yeah. 

Marissa Geist 

I notice you didn't say that the robots run everything, though. In no situation in your mind is AI taking over and we have, like, Skynet. No?  

Calum Chace 

Oh yeah, yeah, that's a separate issue, which I think comes later. So I think we have two singularities, two really big changes coming. One is the end of jobs and the other is superintelligence, which I think probably will come a bit later, although actually they might be quite close together. Superintelligence might follow quite fast. Superintelligence is a machine which has better cognitive abilities than any human in all respects. 

Unless we stop developing smarter and smarter AI, these futures, the end of jobs and the arrival of superintelligence, they are inevitable. I mean, one of two things have to happen for those things not to arrive. One is we stop. We stop making these things. And we're not going to do that because whoever owns the best AI wins whatever competition it is. 

And if you're a company or a government, that's important. If you are military, it's life and death. So we are not going to stop. The other thing that could stop us is if there is some magical ceiling beyond which AIs can’t progress. And I see no reason to believe that. So we are going to get to these stages. 

So yeah, I think when superintelligence arrives, it actually in some senses will run everything inevitably, because it's going to be so much smarter than us. We will become the chimpanzees in this situation. But that's kind of probably further in the future, not the subject you want us to talk about today, which is more the world of jobs and work and so on. 

I'd be interested in your view, Marissa. Do you think there's going to be this future of no jobs? 

Do you think that at some point machines will become so capable that they will replace humans in all jobs? 

Marissa Geist 

I have to say I'm a bit skeptical just because I think I'm limited by my own immediate view. I haven't thought about it as extensively as you have, however, we're seeing that already. We have seen it for years. If we think about just, you know, people that answer phones, people that deliver newspapers, it's not that we found a better way to do something that a human did. 

We actually changed the work, changed the way that people consume, information to consume goods. So we've been on that journey. So I guess it's not that big of a stretch to think that far in the future. I think as we try to pull back and say, well, if that's the future of 2050, let's just accept the premise. There are no jobs. 

We have this utopian society of ultimate, equitable distribution. Sounds amazing. What we're seeing in the workplace are companies really struggling with, well, what should I be doing now then? Does it matter? Should I be skilling people in how to use AI, how to program AI? And we're having a lot of conversations of there will be a finite number of people that really know how to manage and train the AI, and the rest of us will just do more work. 

To your point about not-jobs work. There will just be different ways of working rather than trying to predict the future. Create a very flexible, agile learning organization. Because as incredible as our future seemed in 1985, where we're on this virtual podcast together, I'm interviewing you. I've never met you. You're in a lovely place in Spain. I'm in less lovely but also equally exotic Milwaukee. 

Totally impossible to think that we would do it this way. And if you found the future, you know, if you watch the movies about what people looked like this would be not the same. So I think rather than trying to bet so far out in the future, we're seeing companies look at the skills, the attributes that will still remain the work of humans. 

And that is still up for debate. What is the work of humans if the jobs are taken by AI? And so we're really trying to continue to define that and futureproof the way that we invest and our partners invest as well. But it's, it's difficult to advise. So maybe I'd ask you what is the work that is left for humans if the jobs are all taken by AI? 

What do you see there? 

Calum Chace 

Well, at that point I just think there are no jobs for humans…  

Marissa Geist 

So like no jobs though? Just if I'm thinking… Zero jobs? So someone who cleans my teeth. Someone who does the farming out in a field that's in a very unpredictable terrain. We'll have all of that completely just done. 

Calum Chace 

All done. Even plumbers, even plumbers, which maybe the plumbers may be the last men and women standing. But even them. I mean, there may be some jobs which, you know, you could imagine a job you could devise a job which had to be done by a human. So if you want to have a butler groveling to you, then having a human grovel to you, you might think that's more prestigious than having a robot grovel to you. 

So you could say that job could only be done by a human. But we don't really want lots of humans doing that. 

Marissa Geist 

No, I guess it's like stick shift cars, right? I love stick shift. I drive a stick shift car for the experience, but not because it's actually more practical. Okay, so human choice. Who's training the robots then? Who's training the AI? Is that still a job or? 

Calum Chace 

The robots. They'll train themselves. 

Marissa Geist 

The robots are training themselves. There's actually no Oz behind the curtain, just the robots. Okay. 

Calum Chace 

There might be somebody at the top of each organization or in each country, and perhaps some kind of world government, possibly, where humans make the ultimate decisions about the strategic direction of a company, of an organization or a country. That could be preserved for humans. Although, to be honest, we may well not allow humans to do those sorts of jobs because we just won't be as good at it as a machine. 

You know when they are smarter than us, then they'll make better decisions than us. And that's why we will delegate loads of decisions to them. But one thing that I think people tend to miss is this isn't going to happen gradually, bit by bit. It's going to happen all of a sudden and it's going to happen probably some years into the future. 

And by that I mean maybe 15, 20 years possibly. I mean if Sam Altman is right then it's all gonna happen next year. But probably not. People tend to think that if automation is going to cause lasting, widespread unemployment, then all sectors will disappear one by one over a period of time. As you said, we've had automation for a long time. 

And the classic example is farming, where in 1800, 80% of all Americans who did any work at all worked on farms. In 1900, that was down to 40% and now it’s down to about 1%. And the descendants of those people who are no longer working on those farms, they are doing jobs like the ones we have and working in offices and shops around the world. 

But people didn't become unemployed. People found other jobs to do, and that's going to carry on. Now, a lot of people say that that will carry on forever and will work with computers. We won't be replaced by computers. And that's true, I think, up to a point. And that point is when machines get better at making decisions, at carrying out tasks than we are. When they are more capable than us in every possible respect in the world of jobs, then there won't be jobs for humans. 

Because why would you have a human do a job more expensively and worse than a robot or a piece of software? But I do think that between now and then, there'll be lots of jobs. I think that we'll have more or less full employment until suddenly there's virtually no employment. And I think that change will happen over a few months, because as long as there are some jobs that humans can do and machines can't do, then there's just loads of those jobs for humans, and demand is elastic. 

So there's lots of those jobs. It might end up, I mean, this is a cartoon example, but it might end up with everybody in the world being a plumber because all the other jobs are being done by machines. And plumbing was so hard to organize because it's so tricky and individual. But, you know, eventually even the plumbers go. 

Now, I guess that probably sounds crazy. Does that sound crazy to you? 

Marissa Geist 

It doesn't sound crazy. It sounds probably a little bit too far, but it doesn't. The premise doesn't sound crazy. What I am curious about then is humans have a tendency to then create a revolution of things that they need or want that didn't exist before. So the example of, I can't remember, there's some company that started their business and their call center in an island that had no developed economy before, and everyone made all their money and they made way more money than they did before. 

And then people stopped showing up for work full time because they didn't need the money. They had everything they need. So this company, what do they do? They send them the Sears catalog. And this is like in the 70s. So now you create this whole demand for things you never even knew you needed. You don't actually need in the strictest sense. 

Marissa Geist 

But there's this consumer economy that drives this need to go to work full time when, having a job, actually, it wasn't a part of the society even ten years before that. So AI might create another revolution of a society and a want-based society a greed-based society that we've never seen before. Or will the equitable distribution kind of negate that? 

Calum Chace 

I suspect there will be an equitable distribution, because I suspect the answer to the, what I call the “economic singularity,” which is the point at which there are no jobs for humans, is an abundant economy. And the way I think you get fully automated luxury capitalism is that you drive the cost of everything that you need for a really good standard of living, and I think a continually improving standard of living… 

You drive the cost of everything close to zero. And this is happening anyway, primarily because we automate humans out of more and more bits of the economy, and humans are generally the most expensive part of any production process. We are making energy cheaper and cheaper. You know, it won't be that long before we get… 

I mean, on many days we already get most of our energy directly from the sun rather than by burning dead dinosaurs. And we will, you know, make the rest of that journey over the coming decades, and AI will make all production processes more and more efficient. And so ultimately, you get to a point where the cost of a house, the cost of a car, the cost of food is really, really low. 

And the way that you get fully automated luxury capitalism is probably there is a small minority of people who own most of the assets, but those people get taxed to provide the resources for everybody else. And the taxes are not onerous because the resources it takes to provide for everybody else are not enormous because everything's cheap. So I think that's how you get there. 

And on the way there, I think we all work like crazy because there's loads of jobs for us to do until we get to the tipping point. 

Marissa Geist 

Until we get to the economic singularity. That's an interesting term. I'll have to start using that. So if we now think about today thinking that economic singularity is a possibility, a real possibility on the horizon in the not too distant future. I'm an employer. I run a company, I'm a mother, I have kids, I'm trying to advise. 

I'm also someone who participates in society. What are some things right now that we should be thinking about to make sure that we usher in economic singularity and not dystopian reality? What are some things that just help us get on the right path? 

Calum Chace 

Good question. Well, the first thing is not to get paralyzed by fear that it's all going to a nasty place that it's going to dystopia. Businesses and individuals are faced with today's problems and tomorrow's problems. And today's problems and tomorrow's problems are not enormously different from yesterday's, right now though they are changing. I think probably the first piece of advice that I have is everybody who is not already playing a lot with GPT-4 and its friends, you know, Mistral, Llama and all the others should do so. 

We should all get our hands on AI as much as possible because, it's becoming a cliché but it's true that although for some time an AI isn't going to take your job, another person who works very well with AI may very well take your job. So we all need to become fluent in AI. As users, we don't need to know exactly how a large language model works in fact, nobody knows how much language models work. 

There are black boxes, but we all need to become fluent as users.  

And we all need to think a little bit about the long term as well. It may not be the immediate requirement for businesses and families, because there will be decisions in the transition from our current capitalist economy to fully automated luxury capitalism if we’re lucky enough to get that. 

There’re going to be decisions to make, we're going to need to appoint leaders to make some of those decisions on our behalf. And the more literate we are as a society, as a species, the more likely you will like the decisions.  

I think it's a civic duty to be as up to speed as possible about what AI can do for us now, and where it's going in the future. 

Marissa Geist 

It's really interesting you say that… I sit on the Committee for Economic Development, and a lot of the panels, the webinars that we're on, we talk about the fact that humans have faced changes like this before in the past, and you actually have the playbook of how you think about this. But AI seems so much bigger and so much more scarier than other things we face. 

But we faced a pandemic. Dust off that playbook. You couldn't just say, I'm not doing anything. You have to assess the risk. You have to know what you need to know, and then you have to act in the absence of total information and direct a company. And you can't put your head in the sand, and you also can't try to control things out of your control. 

So we're really encouraging companies to say, if you're talking about it and you're learning about it, you're doing the things that you can do right now. And if you're embracing people that use AI as a part of the way of doing work, and we see the next generation coming in, it's almost like when people used to say, well, you can't use the internet at work because we're not sure it's correct. 

Well, that would be ludicrous now. And actually, we're probably about five years out of people coming into the workforce as the majority of the people coming in expecting to be able to use AI in some format. And we've been using it for years. So make sure you're talking about it. Make sure you're setting up user groups to understand how it's already being used in your business, because it is, whether you know it or not. 

And then really make sure that you're creating the agility to respond to it. And that's what you can do right now. It's be an informed consumer of it and informed leader of it. So that tracks with what we're saying. And go back to remembering that you know how to deal with major change, major shifts. We've lived through once in a lifetime events pretty much every year for the last four years.  

So we're almost, dare I say, experts or it’s a well-worn path for us. So at Cielo, we have literally millions of candidates asking us and being directed by us to enter the workforce every year. And as we look to advise those that are entering the workforce, what are some recommendations you'd make for them with your vantage point of what you think is happening by 2050? 

Calum Chace 

So the good news for young people is that, of course, they're digital natives. They're already very au fait with GPT, they know how to use it better than we do. So they’re going to teach us anyway. But it's the right question to ask. Every talk I gave, I know that everybody in the audience, the question they really want to ask is, what should my son/daughter study at university? 

What career should they go into? How do they futureproof their careers? Now, the bad news is you can't. Because I mean, ultimately the machines are going to take all the jobs anyway, but that's some years away. In the medium term, sadly, we cannot forecast which jobs, which sectors are going to be automated first. All sectors are going to be automated bit by bit, and people are going to have to change to a new sector or change the way they do it. 

And we can't tell what order it's going to be. And we've seen proof of that. Two or three years ago everybody assumed it would be professional drivers, truck drivers, taxi drivers who would be the first to be automated. And hey, it turns out it's translators and people writing not terribly sophisticated marketing copy, and nobody saw that coming. 

And we will be surprised over and over again by, you know, the order in which things get automated, because AI is going to continue to surprise us. So sadly, there's no way to futureproof your career. But I'm confident that jobs are just going to get more interesting because what automation does is it, it has machines taking over the less interesting, the less fun part of a job. 

So young people's jobs are going to be more interesting than jobs we had. They've got these little friends to do the boring bits of the jobs for them, and those friends are going to get smarter and smarter, so it's going to leave the more interesting bits for them. So that's all good.  

But actually, the best advice, the real advice is people should do what interests them. If you pursue your hobby as a career, if you're lucky enough to do that, then you will be doing what you enjoy doing. 

And if you do what you enjoy doing, then you're good at it. And if you're good at it, then you'll succeed.  

Marissa Geist 

That old proverb that said, “Don't do what you like for your work, because then it'll become a job.” 

I guess this flips it on its head, right? If all the non interesting parts of what you do when you like something, you get to give to your little assistant as if you, you know, were an executive in 1980 and having someone trail around taking notes and doing all the things you don't want to do, maybe you can actually now pursue what you like, because there's a really fulfilling career, and you don't have to start to resent it because of all the other stuff that comes with doing what you love. 

So that's a, it's a good twist on what to do now. 

We covered a lot of really interesting topics today, and I have a few key takeaways. One is that we need to embrace AI immediately. It's coming. It's here. It's driving a big change. Two is that we should encourage people to join work that they love, seek out opportunities for what you love to do because work's only going to get better as the AI gets smarter. 

So go for what you love. And the third is don't stop. Don't stop evolving. We can't predict the future. We will have a surprise or two every 2 or 3 years. So be alert, keep your eyes open, but don't be paralyzed by the fact that no one knows what's going to happen. Just keep working. 

I just want to ask if you had one thing to say to someone who is looking for a job or one thing to say to someone who has a team to be thinking about right now, to just make ourselves better prepared for the future. What would it be? 

Calum Chace 

I'd remember that although change has never been so fast, it will never again be so slow. And AI is driving that change, so you really better get up to speed with AI. 

Marissa Geist 

All right, well thank you so much Calum. It was a pleasure to have you on. 

Calum Chace 

Thank you very much. It was a great conversation, lots of fun. And, thank you. 

Marissa Geist 

Join me next time on the trip to work 2050 in The Talent Time Machine.  

To catch the next episode or hear from my previous guests, be sure to follow us on your favorite podcast platform.  

This episode was edited by Matt Covarrubias and produced by Dusty Weis at Podcamp Media... with the support of Sarah Smelik, John McCarron and Laura Pykett of Heavenly and the team at Cielo of Sally Hunter, Annamarie Andrews and Susie Schuppel-Paul.  

For Cielo, thanks for listening... I'm Marissa Geist.