Calum Chace is an author and prolific speaker on artificial intelligence, focusing on how this technology will affect the future of society, culture and the economy. With an academic background in philosophy and thirty years’ experience in business, Chace now devotes his time to spreading awareness about the impending changes technology is to bring us.
I was lucky enough to see him speak at CogX 2018, where his talk on the Economic Singularity was one of the highlights of the event. View my write-up of CogX 2018 to see the talk in full. Following that talk, I was thrilled to have this opportunity to interview Calum Chace about his work and explore his thoughts on the future.
MB: How did you become interested in the issue of AI and how it will affect our future?
CC: I have always been interested in artificial intelligence because I read a lot of science fiction as a kid. My academic training was as a philosopher, and those two things go hand in hand.
Science fiction is an exploration of what it is to be human. It poses questions like, ‘what if the world had no gravity?’ or ‘what if there was no gender?’ and then constructs a world around that. That’s also what philosophy does.
Philosophy tries to figure out the basic questions about truth, belief, about what it is to be human. I always think that sci-fi is philosophy in fancy dress. My interest in those two things run in parallel.
I have always been into AI because it is one of the great tropes of science fiction and a really good intuition pump for philosophy. When I began to realise that we could get to superintelligence, not thousands of years away but within a lifetime or two, it began to get quite urgent to me.
I was in business for thirty years, as a Strategy Consultant and then ran a few companies towards the end. But in 2012, I stopped working to pursue this hobby and then started writing about it. Since then, my hobby has turned into this rather wonderful new career.
MB: And speaking of careers, the main aspect of your work, the thing you are perhaps most well-known for, is your writing on the post-work future. For those who may not already be acquainted with your work, would you give us an introduction?
CC: The argument goes like this. There may be in a generation, not in five years and perhaps not even in ten, a situation in which half the population just cannot get jobs. They just cannot do anything for money which a machine cannot do cheaper and faster. We don’t know if that’s going to happen, and if it is going to happen we don’t know when it’s going to happen, but it’s a real, distinct possibility. There are lots of smart people in the AI world and economists who think it will happen. It would be foolish to just dogmatically say ‘it can’t happen because it hasn’t happened in the past’.
If you accept the argument that it might happen, it’s sensible to think about what we would do if it does. Clearly, if there is a lot of people who do not have jobs, we are going to have to find ways of giving them a good standard of living. When I say ‘them’, I include myself in that. A lot of people jump to the conclusion of universal basic income. But the trouble is that UBI is very expensive. All the schemes that have been tried have essentially failed because it’s too expensive and it’s not enough, even though it’s so expensive.
If half the population is unemployable in a generation’s time, and all they get is a basic income, then society will probably collapse in a very ugly way.
The word in the middle, ‘basic’, is a dead give away. If half the population is unemployable in a generation’s time, and all they get is a basic income, then society will probably collapse in a very ugly way. So the step that most people miss is that the only way to have a massive transfer of income or assets, or both, from those who are still working and those who are rich to those who are no longer doing jobs is if the prices of all the jobs and services that they need for a very good standard of living fall close to zero. That is the ‘economy of abundance’ or ‘Star Trek economy’.
I say ‘close to zero’ because you want to preserve the market. The market is a really good mechanism for telling producers what to produce, what colour, what quantity and where to deliver them to. No other way of organising that has worked. Maybe a superintelligent AI could, but central planning absolutely leads to corruption and dictatorship. So you have got to keep the market, with prices very low.
The economy is dematerialising. A lot of what we want now is digital rather than physical.
Interestingly, that may be where we are going anyway because a lot of the economy is dematerialising. A lot of what we want now is digital rather than physical. A lot of the value is in physical things like cars and mostly computers (cars are computers on wheels now, and that’s just going to become even more the case). So we are heading towards a dematerialised economy anyway.
There has been some interesting work done recently looking at how much Americans consume. They consume more stuff but it costs less to produce the raw materials going into them. We’re becoming much more efficient in our production. We’re heading in that direction but we are probably going to have to accelerate it, and here’s the second step that people miss.
If a massive income or wage transfer is going to be necessary and we haven’t got to the economy of abundance by the time it is necessary, then there is going to be a panic. All the truck drivers and call centre people, all the junior auditors and everybody else who is being laid off will scare the hell out of everybody else. Everybody is going to think, ‘Christ, these machines are coming for my job!’
Now, I don’t know when that panic is going to be. I think it may be in the middle of the time when the truck drivers are being laid off. You can’t fail to see trucks driving around with no drivers in them and I think that will be a big wake-up call, the canary in the coal mine.
What that means is that we need to head off that panic, because if we have that panic then who knows what will happen? It will be ugly. There might be the mother of all house price crashes as everybody thinks to themselves, ‘right, I’ve got to get ahead of this wave, I’m going to convert all my assets into cash so I can withstand this coming shock’. Undoubtedly, some very unpleasant people will get elected, the sort of people who will promise jobs for everybody, promise law and order, and anybody who disagrees will be shot. Oh, and by the way, they’ll say, we’ll shoot all the black people just on principle.
Those sort of people will get elected just because people will be so scared about what’s coming. Those kinds of ‘leaders’ lie, and will be unable to produce all the jobs and security that they claimed, so they’ll have to find an external enemy. The inevitable logic of this is almost certainly war.
There is an alternative, really good, outcome that we can have if we take the issue seriously and confront it properly. We need to recognise that yes, we are heading towards an economy of abundance and there is going to be a bumpy period in between in which prices won’t yet be low enough and we’ll have to have a really souped-up welfare state for all the people who are losing their jobs, or finding a long time to find new ones. This will be temporarily unaffordable, but by the time we get to the economy of abundance that problem will be solved.
We need the economy of abundance and we need to have a plan for it to forestall the panic.
Everybody has to get on board with this over time. Not yet, but increasingly over time, so that there is no need for panic. Joblessness is possible, though we don’t know for sure and we don’t know when. UBI isn’t enough. We need the economy of abundance and we need to have a plan for it to forestall the panic.
MB: In the interim, as you say, there’ll be a lull there before abundance reveals itself. That is perhaps the point at which we are at the highest risk of falling victim to a dictatorship. That seems like almost an inevitability, else there will be riots in the streets. We know that our leaders aren’t fond of the welfare state, being more inclined towards arming the country for or against warfare. How do we keep ourselves politically buoyant?
CC: Well, it’s up to everybody to take this seriously. Politicians, broadly speaking, do what we tell them to do. If they don’t do that, they usually get sacked and replaced with another bunch. We can be as cynical as we like about politicians, but they reflect us and if there is a broad consensus that we’re going to a very good place, to this economy of abundance, then we’ll stand a better chance of success.
Most people don’t enjoy their jobs. I’m sure you do, and I do, but most people out there just do their jobs to put food on the table, or because they believe something better is coming along in the future. A world in which machines do all the jobs and the humans get to do the important business of life, which is learning and playing and exploring and having fun and walking up and down mountains and so on, is a great world. If everybody gets excited about that coming, then I think we can get through the turbulence in the middle.
A world in which machines do all the jobs and the humans get to do the important business of life, which is learning and playing and exploring and having fun and walking up and down mountains and so on, is a great world.
Now, some countries will fail. I confidently predict that, if it carries on the way it’s going that Russia will fail. But if America and Europe and China and other big countries succeed then everybody else will get it and discover that that’s the way to go. It will flow around the world. If goods and services get really cheap, they won’t just be cheap in America, they’ll be cheap everywhere because the knowledge about how to produce stuff cheaply will be there.
MB: How will this impact the super-rich companies though? How will they feel about everything suddenly becoming cheap and abundant?
CC: There will still be relatively rich people and relatively less rich people. If this works out well, however, nobody is going to be dirt poor anymore which will be a huge improvement in humanity, but there will still be people who are much richer than others. Frankly, I don’t think anybody should care much about that. We don’t, as humans, care so much about equality as we do about fairness.
If you watch children playing, they don’t care about one person having more stuff than the other person. They just really hate it if someone gets their extra stuff unjustly. In the short-term and medium-term, inequality like that isn’t going to be a problem. It has to remain that way, otherwise the market won’t be able to function.
MB: But in the long term, as you’ve mentioned in your writings before, there could be a massive polarisation if we get it wrong.
CC: In a sense, I see three phases. The first phase, which we’re not in yet, will be the phase of turbulence as the machines are getting better than us at most things we do. This is the point at we start to see serious, lasting, widespread unemployability. The second phase is of abundance, if we get there. After that, you then have the residual problem of inequality and it gets worse in an interestingly new way.
With technology continuing to accelerate, the technologies which enhance us both physically and cognitively, will be very powerful. We will have neural interfaces, drugs and exoskeletons that make you stronger and cleverer and so on. Initially, those technologies, as is the wont of technology, will be very expensive and won’t work so well, and rich people will be the guinea pigs for them. They will cascade down through the economy and everyone will get to use them, just like smartphones.
As technology accelerates, the rate at which these new technologies become available will be faster and the gap between the people who get them immediately and those who have to wait a month or a year to get them will get sharper. You may, then, get speciation; a polarisation into two or more different species. People who can’t intermarry and can’t even talk to each other, they possibly won’t even be able to understand each other anymore. That is dangerous.
At that point, I’m not sure that a system of ownership, a system of production, is a good idea anymore. The reason why I call this the ‘economic singularity’ is that these changes are so huge that they deserve that term. The other thing about a singularity is that it’s really hard to see what’s on the other side of it. By definition, it’s almost impossible. What that world of the gods and the useless will be like and how we will solve that problem is very hard to tell now.
What that world of the gods and the useless will be like and how we will solve that problem is very hard to tell now.
MB: Even in a fully automated world, there’s still going to be a need – as far as I see it – for humans overseeing the machines and profiting from it.
CC: There’s going to be some humans in jobs for a long time. We will need humans to make strategic decisions about organisations and about countries and so on for a very long time, certainly until superintelligence arrives. So, there will be a minority of people in jobs and being well paid for it. That’s fine, there’s nothing wrong with that as long as everyone else gets to live a good life.
I don’t worry very much about the gods and the useless situation. As Teddy Kennedy once said, let’s drive off that bridge when we come to it. That’s a problem we can park for the future. We’ve got a big enough problem working out whether joblessness is coming and whether the economy of abundance is possible, and how we can get there without too much turbulence. That’s our biggest challenge now and we should focus on that.
MB: We’ve discussed that, in the interim, war may be a distinct possibility, which may be exacerbated by autonomous weapons. I know people are working very hard to keep that regulated, but that doesn’t eliminate the dangers. What are your views on these kinds of weapons?
CC: I have very unpopular views on autonomous weapons. This gets me a lot of criticism amongst those I respect. I think autonomous weapons are inevitable, the reason being that if you have them, you win. Imagine two tanks circling each other in a dogfight. The one that has to phone home to the human in the loop to make a decision, that is the one that loses. Unless you’re happy to be the commander of an army that always loses, you’re going to adopt autonomous weapons sooner or later. If anybody uses them then everybody will.
People say that we have managed to refrain from the use of chemical weapons, and there is an interesting case of light weapons which don’t get used. The difference is, however, that those weapons don’t win you wars, and besides, we haven’t refrained from using chemical weapons! Saddam Hussein used them and Assad uses them. Other people don’t use them because they don’t necessarily win you the war, whereas AI will. Every time there is a battle, whoever uses them will win. I think that makes it even harder to refrain from using them.
Also, I am not convinced that autonomous weapons are really going to be a bad thing. At the moment, the human in the loop is very fallible. I haven’t been in a war myself, but I can begin to imagine what it is like to be a nineteen-year-old, fresh out of the projects in Detroit going through a city in the desert, trying to figure out whether that person crouching in the corner is about to shoot me or whether it’s a woman hiding her child. That’s a position no human should be put in.
Machines will be able to make those decisions better than humans, not least because they will have better visual skills and functionality and faster decision-making capabilities, but also because they don’t care if they get shot. Humans care a lot if they get shot; they get scared and angry. Machines don’t.
I think autonomous weapons may be a lot safer in wars than non-autonomous weapons. I think they may also make warfare a more rational process. One general may reason, ‘they have 300 tanks and we have 150’ and know the outcome. There’s no morale issue, no health and fitness or training issue. We just know that this tank beats that tank every time. Or, if you have two of those tanks, you always win against one of those tanks. So the calculation of war may well become a lot clearer and therefore there may be a lot less of it. Rebels may see that there’s no point going to war against those weapons because they will lose.
At the moment, what happens in war is that old men send young men to die. I think it would be better if old men sent machines to die.
The terrifying thing about autonomous weapons is if they go wrong. They will go wrong and there will be accidents. But we put a lot of faith in quite advanced technology already. Most of us don’t know how our cars work. I certainly don’t. We are, however, happy to drive along roads with other people driving two tonne vehicles within inches of us for hours. Aeroplanes don’t routinely fall out of the sky, nuclear arsenals don’t routinely get taken over by hackers, and nuclear power plants don’t routinely blow up. So, there’s no reason to think that autonomous weapons are going to be that much more unreliable.
At the moment, what happens in war is that old men send young men to die. I think it would be better if old men sent machines to die.
MB: I’d just like to shift a little to talk a bit about the technological singularity. Obviously, this is a few years along from the economic singularity, but certainly the two are linked in some ways. Elon Musk has famously said that we will need to enhance ourselves if we are to survive as a species, that we need to evolve to keep up with the AI should the technological singularity occur.
CC: I am absolutely in awe of Elon Musk. He’s an extraordinary guy and he does extraordinary things. One of his great skills is that he thinks very clearly and is willing to accept conclusions from a line of argument that most people will just shy away from.
His argument here was that superintelligence is probably coming and it will be either the best or the worst thing that’s ever happened to us. He does think it could be the best thing to happen to us, he’s not entirely a Cassandra. Obviously, however, there are risks if we’re not going to take it seriously, to prepare for it and make sure that the first one to arrive really does like us. Musk has been banging that drum for a while and he thinks nobody is listening, so he changes his tack. If we’re not going to do it that way, he thinks, then maybe we try to make sure that they don’t overtake us.
I think he’s wrong about that, that it’s not possible, and certainly not in the timeframe. But his Neuralink venture will, I’m sure, throw up some really useful technologies anyway so it’s all to the good.
I just wish that Musk, who I think is very brave and a great spokesperson for AI issues, would speak more about the economic singularity. He’s been very thin on that. He’s tweeted a few times but he seems to be of the opinion that universal basic income will just take care of it, and I disagree.
Having said that, the technological singularity is bigger. The economic singularity, if it goes wrong, may just throw us back to the middle ages. We can recover from that. But if we get the technological singularity wrong, if the first superintelligence to arrive on the planet has a deep hatred for humans, then we’re stuffed. So, it is right for him to bang that drum.
The good news is that there are four organisations now, and there will be more, that are taking that project seriously and addressing it. They’ve also probably got a few decades to sort it out, so I don’t worry so much about that, though I do agree that it’s a very important project.
MB: So should we enhance ourselves, as Musk suggests?
CC: Whenever a technology satisfies a genuine need, it will be used. The more compelling the use case, the quicker and more widespread its use will be. In pretty much all walks of life, I doubt that we will be able to augment ourselves as quickly as machines will take over our roles, wherever that role is functional. Machines will not take over our human role, which is, as I say, learning, playing, exploring, falling in love and the rest of it. They can’t take those because they are not conscious. So the really important stuff, we will preserve and machines will just do the boring stuff.
Machines will not take over our human role, which is, as I say, learning, playing, exploring, falling in love and the rest of it. They can’t take those because they are not conscious. So the really important stuff, we will preserve and machines will just do the boring stuff.
Augmenting ourselves probably isn’t the way to stop them. We’ll do it anyway, because an augmented human is, in many ways, better than a standard human. I really relish the idea that I could have a new heart when mine fails, keep swapping bits out as they fail, and if I could have much better vision, sense of smell, be able to leap over tall buildings, that would be great!
MB: Coming back to the present, how should organisations begin preparing for mass automation?
CC: I don’t think it’s the job of commercial organisations to try and sort out the economic singularity and ensure it works. People who run those companies are often pretty wealthy and influential, so they should be thinking about it and perhaps allocating some money to think tanks. But it’s not the company’s job to sort it out.
What companies should be doing is to deploy AI intelligently. It is very early days. The tech giants are making stupendous amounts of money with deep learning and very few other people are, because it’s hard and not always immediately obvious what the right way to do things is, and what the right order is. That’s a big enough challenge, figuring out deployment.
As for how they should treat their workforce as and when they let them go, the more humane and considerate and helpful they can be, the better. That’s a moral issue rather than a commercial issue and I do think that companies should have in mind that, as unpopular as it may be, the primary job of the company is to increase value. It should do so within the law and do so morally, but that is its primary job and that’s how the market works. Companies that are constructive and helpful to those they let go, helping them to retrain and to find new jobs, to redeploy them where they can, will benefit from the fact that people will want to work for them and customers will like them. So there is an enlightened self-interest in being helpful towards your staff.
The job of working out how to navigate the economic singularity is the job of politicians and policy makers. But it is also the job of all of us, to get the politicians to wake up and start thinking about it, to demand they do so, and also to enable the public to vote about it and to be well-informed about it by informing ourselves. The best way to keep educated is, of course, through reading. You could start with my book!!
MB: Let’s also touch on VR. I know you’ve said before that you see virtual reality as having a really significant role in the future. I imagine that as a kind of Ready Player One scenario when placed into the environment of the jobless future. But perhaps you have another angle on that?
CC: I strongly suspect that VR will play a very important role in all of our lives. I don’t think it’ll take many people over and consume them. It may happen to some people, just as some people today fall into drugs and drink and other mindless pursuits. But most people will treat VR in the same way that we treat television. It’ll be an important aspect of our lives and in many ways, that’s a good thing. VR is pretty resource-light, you don’t need lots of bits of metal and so on to access it.
Humans are what we are because of how evolution has shaped us. And it has shaped us to be really good at surviving on the savannas of Africa. We spent a million years or so learning to be really good at that. One of the things that makes us the apex predator is our sociability. We collaborate, and it is only by collaborating that we were able to survive attacks from tigers, to kill mammoths and build civilisations. That conditioning isn’t going to go away in a generation. People will always want to talk to each other and socialise. I don’t see that VR will turn us all into sad losers isolated from the world.
It’s not what happens now. I have a teenage son and he likes videogames. He doesn’t play alone, he plays with other people. Some online, some in the same room. I don’t see why we would stop being social all of a sudden.
In this world, it is just not possible for everyone to have an original Vermeer or to drive an original Aston Martin DB5. But in virtual, and perhaps augmented, reality, everybody can. They can have something very close to the actual experience of it, which will be better than not having something very close to the actual experience of it!
It could make all of our lives richer.
MB: Do you see education becoming very VR-centric, and tying that into the issue of the jobless future, how do you see the evolution of the education system?
CC: I think education will deploy VR and AR. It’d be crazy if it didn’t; it’s a great tool. I also think education is going to get transformed within the next twenty years in ways which teachers have no clue about.
We will have things that people call ‘edubots’, AIs that are our personal tutors that are with us throughout our lives. This AI will know us very well, what we need to know next in the curriculum and it knows what we already know. It also knows how we learn best and how to encourage us. I think it will be phenomenally effective, much more so when we have these things. They will use virtual reality and every other tool that they can lay their hands on.
Education, and life at large perhaps, needs to become vacational rather than vocational.
A lot of people believe that the education is very resistant to change, but it actually doesn’t do a terribly bad job. A lot of the people who berate the education system were, in fact, educated by it. They wouldn’t consider themselves to be terribly poorly educated and they are the products of it.
I think that things will change as the technology becomes available, not because somebody decides how they’re going to change education. The education system, not just in this country but everywhere, is really good at resisting change but when the tech comes along that allows you to have your own, personal educator, they will not be able to resist that.
MB: If we move into this post-work future, what are we educating for?
CC: To enjoy our lives more. If you have a life of leisure, but you are ignorant, then your life isn’t as rich as if you know a lot about the way the world works and what the history of your surroundings is. If you climb a mountain and you know lots about the geology, how the mountain was formed, about the flowers and birds, if you know the history, about the armies that have fought over this land, then you have a much richer experience than if you don’t have a clue what you’re looking at. I think education is just part of life. It’s a shame to see it as simply a tool to get you a better job.
Education, and life at large perhaps, needs to become vacational rather than vocational.
MB: So what would you do in a post-work society?
CC: I think pretty much what I do now, which is to speculate about what’s happening next and discuss it with interesting people. If I think there’s an important lesson or issue that people need to think about, then I would agitate and lobby for everyone else to think about it.
All of Calum Chace’s books are available to buy on Amazon. You can also receive a free copy of his book, Our Jobless Future, by signing up to receive emails via his website here.