Artificial intelligence is expected to create more value for global businesses than the current worth of the entire United States economy over the next 15 years.
That figure—$19.2 trillion—comes from an economic model commissioned for a new book called The AI Age by Adam Riccoboni, managing director at UK-based tech consultancy Critical Future. The book gives an overview of the various ways in which machine learning and automation research is expected to transform society in coming years, ranging from job market flux to the fundamental operations of various industries.
The book also claims to be the first to feature cover art generated by AI. More specifically, it’s a cutting-edge type of machine learning system called a generative adversarial network (GAN) that’s trained on a database of 200,000 real book covers.
Riccoboni spoke to Adweek about why he remains optimistic about the future of AI overall, what workers should do to prepare for an AI economy and what its most exciting uses are.
This interview has been condensed and edited for clarity.
Adweek: What is the significance of this $19 trillion figure you’ve calculated?
Adam Riccoboni: It’s a huge amount. That’s about the same size of the U.S. economy, basically. And that is now going to be distributed to businesses that invest in AI, and people that focus their careers in the right way. So there’s this great boon for some people, but the other side, there’s going to be a lot of disruption for other people. Millions of jobs are going to be disrupted, millions of other jobs will be created.
But people need to think about how to get on the right side of this big technological revolution. Imagine if before the internet happened, you could see the internet was going to come, you knew publishing was going to be ripped apart, that you might have to become a digital marketer instead of just a marketer or a programmer or a social media manager—all these different changes that the internet brought in. Something similar is going to happen over the next 10 to 15 years of artificial intelligence. And people need to be prepared.
There are a range of views among experts, from optimistic to dystopian, on whether economic disruption caused by AI will ultimately be a net plus for humanity. Where do you stand on that spectrum?
I’m an optimist. I definitely think there will be millions of jobs displaced. For example, originally, these two Oxford academics caused a bit of a fray, when they found 47% of jobs in the U.S. would be displaced. That’s come down to about 9% now in the [Organisation for Economic Cooperation and Development (OECD) member countries,] but there’s still millions of jobs here. The thing is, millions of other jobs will be created. The important thing is that it changes the balance of jobs.
You have to think, “Where is a human’s comparative advantage over machines? What are we better at than machines?” Machines are better at doing calculations, doing repetitive things, all these type of machine jobs that will be taken over by machines. And then we will be freed up to do things like build relationships, more creative jobs, more general problem-solving skills. That’s our advantage over machines is we’ve got general problem solving skills; they can only ever do very narrow technical things.
You just have to be very adaptable. So because of that, you should have a well-rounded education, you should have an interdisciplinary education. You should read history, politics, philosophy, not just the technical things—all of those things to give you a well-rounded set of general problem-solving skills to make you adaptable, make you strategic, make you creative, and then you’re going to be doing very well in the transition.
What are some of the specific applications of AI that you see as most promising or exciting at the moment?
Andrew Ng said that deep learning is like the discovery of electricity. So essentially, this hugely important technology has been discovered, and now it’s in the implementation phase. And there’s just endless opportunities. At our company Critical Future, we’ve got lots of PhDs in machine learning from Harvard, from Cambridge University, really top guys. And we’re working on rolling this technology out in all different use cases.
For example, we’ve worked on predicting property prices from a real estate firm using deep learning. In two months, we outperformed all their models they’ve been working on for eight years. We’ve also predicted cancer with 100% recall accuracy from a photo of a mole on your skin for a healthcare company. We just did a project to predict stock prices and commodity prices—we were predicting the price of a metal—and we got better results than a hedge upon that we benchmark against him in New York. So there’s there’s endless opportunities. Really it’s just a question of companies having to think about what’s valuable to predict in their industry, and you can use deep learning for that.
I also do think GANs are a really exciting field because they show machines can be creative. A machine can create a piece of art or a piece of music as well. And what they actually do is they’re creating an amalgamation of what they’ve learned in their training. So for example, when we created our book cover, that book cover was actually an amalgamation of the 200,000 books that have gone in.
So you could say, “Is that real creativity? You know, it’s just an amalgamation.” But I would say, yes, it is. Because we do something similar as human beings. We take in lots of data, and then we recombine it to create something new.