In part 1, futurist Byron Reese told us what to expect from automation and AI in the future. Here he flips the story to tell us what we shouldn’t expect. In other words, what can people do that technology can’t?
Technology is only a tool
AI is a tool that can enhance human productivity by creating a future that’s like the past. So if that’s what we want – to do the same thing we’ve been doing, only to do it more efficiently, that’s fine. We can do that with AI.
“The way we do AI now is with machine learning,” Reese explains, “where we take data about the past and study it and make projections based on it. So by definition it can only do tasks where the future is like the past.”
Which means there can be no unknown unknowns. Which means no creativity. Which means no invention. Which means no airplane. No landing on the moon. No iPhone. No Guernica. No Rapper’s Delight. No Facebook.
“I firmly believe that if a machine can, in theory, do a job, then to make a human do that job is dehumanizing. A machine’s not creative. A machine’s not inspirational. A machine’s not passionate.”
“Humans are. You are. If the work doesn’t need creativity or inspiration or passion, then we don’t need you to do it. If a machine can do a job, a machine should be doing it. A machine can do any job where two people would do it the same way, like data entry, where two days of the job are virtually identical, and where it takes less than one second to do a cognitive task.”
“When I look out my window right now, I can tell you a dozen things that need doing that only people can do. There’s a wall that needs a person to paint a mural. There's a median that needs beautifying – that needs flowers. Machines can’t do that. There’s no limit to what only humans can do.”
“The urgency is to build machines to do the things that humans don’t want to do. I don’t think most marketing people would want to write the kind of copy that computers can write. It’s lifeless. It’s soulless. It’s as dead as fried chicken.”
A computer can't get creative
Then the question becomes — for those of us who are worried that computers might take over our jobs, what is that spark that separates the computers from us? If a computer can write marketing copy, can it write a novel?
What is that spark that separates the computers from us? If a computer can write marketing copy, can it write a novel?
Reese laughs. “The people in my podcast — the people who believe we are machines — laugh at that. Of course a robot can write a novel! They view that as a simple task. But this is still an open question in my mind. I don’t think they can create something completely new. Of course, the guests on my show would say humans can’t either. If what you do with AI is take information about the past and project it into the future, if you study and look for patterns, I don't know how you get JK Rowling out of that. But again, many AI researchers vehemently disagree with that statement.”
“Lin-Manuel Miranda and Hamilton. Banksy. I think there is something – the X-factor – that we don't understand about people. There are different choices of what it could be. Some people believe they're animated by their soul and from it, all the creativity flows. Other’s believe our creativity is a kind of strong emergent property that is beyond our present science, but that it is simply science.”
We're more than the sum of our parts
Even if we could completely replicate a human brain with a computer program, it still might not work the way a human brain does.
As Reese explains, some people think that little bit extra is their soul. Some people think it’s emergence. (Yet another concept I learned that I plan to drop casually at the next cocktail party I attend so I will sound smart.)
“Emergence is a property whereby a whole takes on characteristics that none of the parts have.”
“One example of emergence would be the sense of humor. You're made of a trillion cells, none of which have sense of humor, and yet you have a sense of humor. Whence did it come? The idea is that it's emergent, that somehow it's the interplay of all of those.”
“We don’t have a good scientific answer to this question is because we don't understand the mind. You have a hundred billion neurons in your brain that are connected to each other and do all these amazing things that no neuron can do.”
“But that isn't the reason we don't understand it. It's not because there are so many neurons.”
“There's a twenty-year-old effort called OpenWorm. The nematode worm has only 302 neurons in its brain, and it does all kinds of pretty sophisticated things with these 302 neurons.”
“And there are people who have spent 20 years trying to figure out, ‘How can we arrange these 302 neurons and the connections between them and get it to behave the way a nematode worm does?’ They are trying to model that in a computer, to get what would essentially be a new life form, a digital nematode worm.”
“And not only have they not done it, but there’s debate whether it’s even possible. We don’t know how a thought is encoded in the brain, in the mind.”
“If I were to ask you, ‘What color was your bicycle as a kid?’ you could probably answer and answer easily. Mine was green. Yet there’s not a place in the brain that stores bicycle information. Nobody knows how it happens that you can recall that. Nobody knows where that information is encoded in the brain. Well, maybe somebody knows, but they sure aren’t telling anyone.”
“That's why we can't build machines to do with what people can do: we don't have a clue how we do what we do.”
We can stretch our skill set
OK, so a machine can’t replace me completely because of context and emergence and all that stuff. But there is automation and AI that can do part of my job. So what do I do? I don’t work as a hobby.
In the Industrial Revolution, we automated and mechanized and eventually made life better for everybody. But there was this social upheaval that was really awful for a lot of people.
We have that problem today. How do we address that? I hear people saying, nope, you can’t mine coal anymore, but you can become a computer programmer. Which — I don’t see that happening.
Reese, who, in addition to being a futurist, is also an economist, has an answer.
“I've spent a lot of time trying to figure out the half-life of a job, and I think it's 50 years. I think every 50 years, we lose half of all jobs. But not only have we not had systemic unemployment in the last 250 years, we have had ever higher wages.”
“I think it works like this. Technology is really good at making these new high-end jobs, like geneticist. But it destroys all these jobs at the bottom, like order taker for fast food. And then people ask, ‘Are you really going to turn that order taker into a geneticist?’”
“No! That’s not what happens. What happens is the college biology professor becomes the geneticist. And the high-school biology teacher gets the college job. The substitute teacher is hired on full time. And all the way down the line.”
“So the question is not, ‘Can that coal miner code?’ The question is, ‘Can everybody do a job that’s a little bit harder than the job they have today?’"
So the question is not, ‘Can that coal miner code?’ The question is, ‘Can everybody do a job that’s a little bit harder than the job they have today?
“And if the answer to that is, ‘Yes,’ which I absolutely think it is, then that’s 250 years of economic history, which is, you destroy jobs at the bottom, you create jobs at the top, and everybody shifts up.”
We can adapt to new opportunities
Reese corrects me. “But the Industrial Revolution may not be the example to use. I actually think it's the internet because it's only 25 years old. So if we went back 25 years and said hey, here's the internet. In the future, billions of people are going to use this. What's it going to do to jobs?”
“If you were really smart, you would say no more stockbrokers, no more phone books, no more travel agents. Newspapers are going to have a hard time.”
“And you would have been right about everything. You would see all the things the internet is going to destroy. But nobody saw eBay or etsy or AirBNB or google or $25 trillion in wealth created by the internet.”
“And so I would ask the question this way: has the internet been good for workers or bad for workers?”
It takes me a nanosecond to answer.
I am working at home today specifically because I do not have to be physically in the same room with my coworkers to have a meeting or to review a report or, actually, to do anything required to do my job.
Reese agrees. “This new technology came out. It wiped out a lot of jobs and changed a lot of things. A bunch of people lost their jobs.”
“But the great skill people have is the ability to learn new things.”
“I went to high school in the mid-80s. There’s only one class I took then that’s useful today: typing. Everything else I know, everything I do in my job, I taught myself. People are uniquely able to do that.”
We can adapt to new challenges
I note that a lot of the benefits of new technologies seem to help employers more than they help employees. I am now expected to attend meetings that start at 5:00 a.m. Yes, I can attend them from home in my PJs with a cat in my lap, but it’s still 5:00 a.m.
Reese agrees that much of the benefit of technology accrues to the owners of the labor. “Yes. If you own your labor, if you're a lawyer and you write wills and you get some new technology that helps you write four wills in the time it used to take you to do five, you get that money.”
“But on the other side of the equation, if you’re a cashier at a discount store and sell your time by the hour and your company puts in new technology that doubles your productivity, you don’t get any of that. It’s not a big conspiracy; it’s just how the chips happen to fall in our system. If we want to change that, well, that’s a policy question.”
“But. How much money would it take for you to never use the internet again?”
How much money would it take for you to never use the internet again?
Ah. Well. That's a different question
And an easy one to answer: you will take my smartphone out of my cold dead hands. I hold more power in my hand than NASA used to send the first humans to talk on the moon. I can get information whenever I want.
Reese the economist notes, “And that doesn't show up in any economic numbers. You didn’t used to have that ability but now you do. In terms of your wealth — of everything you have — you are vastly better off than you were.”
“When you hear people say frightening things about AI, they say them because they think they’re machines. But if you don’t think you’re a machine, you have nothing to fear.”
Related Posts:
