A.I economics: How cheaper predictions will change the world

Predicting the future is about to become a whole lot cheaper. Here's how economists look at artificial intelligence.

Ajay Agrawal: I think economics has something to contribute in terms of our understanding of artificial intelligence because it gives us a different view. So, for example, if you ask a technologist to tell you about the rise of semiconductors they will talk to you about the increasing number of transistors on a chip and all the science underlying the ability to keep doubling the number of transistors every 18 months or so. But if you ask an economist to describe to you the rise of semiconductors they won’t talk about transistors on a chip, instead they’ll talk about a drop in the cost of arithmetic. They’ll say, what’s so powerful about semiconductors is they substantially reduced the cost of arithmetic.

It’s the same with A.I., everybody is fascinated with all the magical things A.I. can do and what economists bring to the conversation is that they are able to look at a fascinating technology like artificial intelligence and strip all the fun and wizardry out of it and reduce A.I. down to a single question, which is, “What does this technology reduce the cost of?” And in the case of A.I. the recent economists think it’s such a foundational technology and why it’s so important it stands in a different category from virtually every other domain of technology that we see today, is because the thing for which it drops the cost is such a foundational input, we use it for so many things; in the case of A.I., that’s prediction.

And so why that’s useful is that as soon as we think of A.I. as a drop in the cost of prediction, first of all, it takes away all the confusion of well, what is this current renaissance in A.I. actually doing? Is it Westworld? Is it C-3PO? Is it a Hal, what is it? And really what it is, it’s simply a drop in the cost of prediction. And we define prediction as taking information you have to generate information you don’t have. So it’s not just through the traditional form of forecasting like taking last months sales and predicting next months sales. It’s also taking, for example, if we have a medical image and we’re looking at a tumor and the data we have is the image and what we don’t have is the classification of the tumor as benign or malignant, the A.I. makes that classification, that’s a form of prediction. And so when something becomes cheap—from economics 101 most people remember there’s a downward sloping demand curve—and so when something becomes cheaper that means we use more of it. And so in the case of prediction as it becomes cheaper we’ll use more and more of it. And so that will take two forms: one is that we’ll use more of it for things we traditionally use prediction for like demand forecasting and supply chain management. But where I think it’s really interesting is that when it becomes cheap, we’ll start using it for things that weren’t traditionally prediction problems but we’ll start converting problems into prediction problems to take advantage of the new, cheap prediction.

So one example is driving. We’ve had autonomous cars for a long time, or autonomous vehicles, but we’ve always used them inside a controlled environment like a factory or a warehouse. And we did that because we had to control the number of—think of it as the if/then statement. So we have a robot, the engineer would program the robot to move around the factory or the warehouse and then they would give it a bit of intelligence. They would put a camera on the front of the robot and they would give it some logic, saying okay if something walks in front then stop. If the shelf is empty then move to the next shelf. If/then. If/then.

But you could never put that vehicle on a city street because there is an infinite number of ifs. There are so many things that could happen in an uncontrolled environment. That’s why as recently as six years ago experts in the field were saying we’ll never have a driverless car on a city street in our lifetime—until it was converted into a prediction problem. And the people who are familiar with this new, cheap form of prediction said why don’t we solve this problem in a different way and instead we’ll treat it as a single prediction problem? And the prediction is: What would a good human driver do?

And so effectively the way you can think about it is that we put humans in a car and we told them to drive and humans have data coming in through the cameras on our face and the microphones on the side of our heads and our data came in, we process the data with our monkey brains and then we take action. And our actions are very limited: we can turn left; we can turn right; we can brake; we can accelerate. The way you can think about it is, think about an A.I. sitting in the car along with the driver and what that A..I is trying to do is—it doesn’t have its own input sensors, eyes and ears, so we have to give it some: we put a radar camera, LiDAR, around the car—and then the A.I. has this incoming data and every second it’s got data coming in, it tries to predict in the next second what will the human driver do? In the beginning, it’s a terrible predictor it makes lots of mistakes. And from a statistical point of view, we can say it has big confidence intervals; it’s not very confident. But it learns as it goes and every time it makes a mistake, it thinks that the driver is about to turn left but the driver doesn’t turn left and it updates its model. It thinks the driver was going to brake, the driver doesn’t brake, it updates its model. And as it goes, the predictions get better and better and better and the confidence intervals get smaller and smaller and smaller.

So we turned driving into a prediction problem. We’ve turned translation into a prediction problem. That used to be a rules-based problem where we had linguists with many rules and many exceptions and that’s how we did translation. Now we’ve turned it into a prediction problem.

I think probably the most common surprise that people have is we have a lot of HR people that come into our lab and they say: 'Hey, we’re here to learn about A.I. because we need to know what kinds of people to hire for our company you know, for our manufacturing or our sales or this or that division. Of course, it won’t affect my division because I’m in HR and we’re a very people-part of the business and so A.I. is not going to affect us.' But of course, people are breaking HR down to a series of prediction problems.

So for example, the first thing HR people do is recruit, and recruit is essentially they take in a set of input data like resumes and interview transcripts and then they try to predict from a set of applicants who will be the best for this job. And once they hire people then the next part is promotion. Promotion has also been converted into a prediction problem. You have a set of people working in the company and you have to predict who will be the best at the next-level-up job. And then the next role they do is retention. They have 10,000 people working in the company and they have to predict which of those people are most likely to leave, particularly their stars, and also predict: what can we do that would most likely increase the chance of them staying? And so one of the, what I would say, a black art right now in A.I. is converting existing problems into prediction problems so that A.I.s can handle them.

When most of us look at A.I. we see magical capabilities. When economists look at A.I. they see something very different. Economist Ajay Agrawal explains: "What economists bring to the conversation is that they are able to look at a fascinating technology like artificial intelligence and strip all the fun and wizardry out of it and reduce A.I. down to a single question, which is, 'What does this technology reduce the cost of?'" Never has one person taken such delight in stripping the fun from something awesome. But what does A.I. lower the cost of? Predictions, says Agrawal. Intelligent machines can take information we have and use it to generate information we need. Uncertainty is the single biggest hurdle in good decision making, and A.I. can drastically increase certainty in many areas, like automated vehicles, language translation, human resources and medical diagnostics. As A.I. becomes a cheaper technology, its use will become even more widespread. "Where I think it’s really interesting is that when it becomes cheap, we’ll start using it for things that weren’t traditionally prediction problems but we’ll start converting problems into prediction problems to take advantage of the new, cheap prediction." Ajay Agrawal is the co-author of Prediction Machines: The Simple Economics of Artificial Intelligence.

Global climate strike: Scenes from the #ClimateMarch protests

The week-long global protest, which is calling for an end to the age of fossil fuels, is taking place in more than 160 countries today.


SOPA Images
/ Contributor / Getty
Politics & Current Affairs
  • Millions of people around the world are taking to the streets to demand more urgent action on climate change.
  • The protests come just days ahead of the 2019 UN Climate Action Summit.
  • Although it's unclear exactly how many people are participating, it's likely to be the largest climate protest ever.
Keep reading Show less

How do 80-year-old 'super-agers' have the brains of 20-somethings?

Most elderly individuals' brains degrade over time, but some match — or even outperform — younger individuals on cognitive tests.

Mind & Brain
  • "Super-agers" seem to escape the decline in cognitive function that affects most of the elderly population.
  • New research suggests this is because of higher functional connectivity in key brain networks.
  • It's not clear what the specific reason for this is, but research has uncovered several activities that encourage greater brain health in old age.

At some point in our 20s or 30s, something starts to change in our brains. They begin to shrink a little bit. The myelin that insulates our nerves begins to lose some of its integrity. Fewer and fewer chemical messages get sent as our brains make fewer neurotransmitters.

As we get older, these processes increase. Brain weight decreases by about 5 percent per decade after 40. The frontal lobe and hippocampus — areas related to memory encoding — begin to shrink mainly around 60 or 70. But this is just an unfortunate reality; you can't always be young, and things will begin to break down eventually. That's part of the reason why some individuals think that we should all hope for a life that ends by 75, before the worst effects of time sink in.

But this might be a touch premature. Some lucky individuals seem to resist these destructive forces working on our brains. In cognitive tests, these 80-year-old "super-agers" perform just as well as individuals in their 20s.

Just as sharp as the whippersnappers

To find out what's behind the phenomenon of super-agers, researchers conducted a study examining the brains and cognitive performances of two groups: 41 young adults between the ages of 18 and 35 and 40 older adults between the ages of 60 and 80.

First, the researchers administered a series of cognitive tests, like the California Verbal Learning Test (CVLT) and the Trail Making Test (TMT). Seventeen members of the older group scored at or above the mean scores of the younger group. That is, these 17 could be considered super-agers, performing at the same level as the younger study participants. Aside from these individuals, members of the older group tended to perform less well on the cognitive tests. Then, the researchers scanned all participants' brains in an fMRI, paying special attention to two portions of the brain: the default mode network and the salience network.

The default mode network is, as its name might suggest, a series of brain regions that are active by default — when we're not engaged in a task, they tend to show higher levels of activity. It also appears to be very related to thinking about one's self, thinking about others, as well as aspects of memory and thinking about the future.

The salience network is another network of brain regions, so named because it appears deeply linked to detecting and integrating salient emotional and sensory stimuli. (In neuroscience, saliency refers to how much an item "sticks out"). Both of these networks are also extremely important to overall cognitive function, and in super-agers, the activity in these networks was more coordinated than in their peers.

Default Mode Network

Wikimedia Commons

An image of the brain highlighting the regions associated with the default mode network.

How to ensure brain health in old age

While prior research has identified some genetic influences on how "gracefully" the brain ages, there are likely activities that can encourage brain health. "We hope to identify things we can prescribe for people that would help them be more like a superager," said Bradford Dickerson, one of the researchers in this study, in a statement. "It's not as likely to be a pill as more likely to be recommendations for lifestyle, diet, and exercise. That's one of the long-term goals of this study — to try to help people become superagers if they want to."

To date, there is some preliminary evidence of ways that you can keep your brain younger longer. For instance, more education and a cognitively demanding job predicts having higher cognitive abilities in old age. Generally speaking, the adage of "use it or lose it" appears to hold true; having a cognitively active lifestyle helps to protect your brain in old age. So, it might be tempting to fill your golden years with beer and reruns of CSI, but it's unlikely to help you keep your edge.

Aside from these intuitive ways to keep your brain healthy, regular exercise appears to boost cognitive health in old age, as Dickinson mentioned. Diet is also a protective factor, especially for diets delivering omega-3 fatty acids (which can be found in fish oil), polyphenols (found in dark chocolate!), vitamin D (egg yolks and sunlight), and the B vitamins (meat, eggs, and legumes). There's also evidence that having a healthy social life in old age can protect against cognitive decline.

For many, the physical decline associated with old age is an expected side effect of a life well-lived. But the idea that our intellect will also degrade can be a much scarier reality. Fortunately, the existence of super-agers shows that at the very least, we don't have to accept cognitive decline without a fight.


Millennials and the rise of tiny homes

Are tiny homes just a trend for wealthy minimalists or an economic necessity for the growing poor?

Photo credit: Cyrus McCrimmon / The Denver Post via Getty Images
Politics & Current Affairs
  • The tiny home movement has been popular on social media sites, often portraying an idyllic lifestyle that's cheaper and better for the environment without sacrificing aesthetics.
  • But tiny homes may become the answer to a growing population and growing inequality.
  • As the movement continues to build up steam, one has to wonder whether it's a housing crisis solution with a new coat of paint.

Tiny homes. They're the watchword of the Home & Garden network, at once an Instagrammable, envy-inducing lifestyle and an unfortunate necessity for a generation struck by a recession, historically high inequality, and loans taken out for an ostensibly necessary education that's failed to really net any benefits.

But the question is, which are they? A symbol of a smarter, more environmentally-conscious, humbler generation — or a symbol of one that's had to make do with less than its predecessors? (See: "Millennials buy the things their parents did — but they're much poorer.")

Downsizing housing and hubris

Image source: Mike Morgan / For The Washington Post via Getty Images

Will tiny homes look like this in the future -- smaller and more efficient but still beautiful?

In the U.S., things are just bigger, and houses are no exception. The median size of a single-family home in the U.S. peaked in 2015 at 2,467 square feet. Compared to other parts of the world — particularly Europe — this is a massive figure. There's a variety of reasons for this; one, for example, is that Americans began driving early and often, which transformed the design of their cities and suburbs. Developers could build outside of urban centers where the land was cheaper and more plentiful, enabling bigger houses to be bought.

In addition, the idea of having a lot of space seems to be an appealing one to the former European colonies — where Europeans have often lived in more cramped, repurposed older buildings, Australians, Canadians, and Americans had the opportunity to seize land (despite it already being occupied) and build new, sprawling settlements throughout it. The prosperity that the America saw in the 20th century didn't hurt, either; why not build big if you've got the money to spare?

But a considerable amount of this space is wasted. A UCLA study found that the majority of people spend their time in the kitchen or around the television and very rarely use the living room or porch. As a result of these extra, unused spaces, more resources are wasted on construction, and energy consumption is double what a family would need if their house only had the rooms that they actually use.

Smaller, more energy-efficient houses are appealing to a growing population of minimalists and resource-conscious individuals. In 2017 alone, the sales of tiny homes increased by 67 percent. Coming in at under 400 square feet on average, these houses are also understandably cheap — for tiny homes on wheels, the average cost is $46,300, while those with a foundation cost on average $119,000. As a result, 68 percent of tiny homeowners don't even have a mortgage.

Downsizing out of necessity

Tiny homes

Image source: George Rose/Getty Images

A community of tiny homes for homeless people known as "Nickelsville" in Seattle.

On the other hand, the group of people drawn to tiny homes isn't just homogenously composed of wealthy minimalists looking to reduce their consumption while still appearing trendy. In 70 percent of the U.S., the average worker can't afford a home, one-third of adults are a $400 bill away from financial difficulty, and a quarter have no retirement savings whatsoever.

Under these conditions, downsizing may be the only viable method to survive. Consider, for instance, how cities such as Seattle, Detroit, and Denver are constructing tiny homes as emergency shelters or transitional housing for the homeless. There are also the many retirees that had their savings wiped out by the Great Recession who now live nomadically in RVs and modified vans. This tiny-living trend also has its Instagram cheerleaders, but the reality of it is less idyllic. Journalist Jessica Bruder and author of Nomadland related an anecdote to MarketWatch illustrating the nature of nomadic tiny living:

"I talked to one couple, Barb and Chuck. He had been head of product development at McDonald's before he retired. He lost his nest egg in the 2008 crash and Barb did, too. One time, Barb and Chuck were standing at the gas station to get $175 worth of gas and the horror hit them that their account had $6 in it. The gas station gentleman said 'Give me your name and driver's license and if you write a check, I will wait to cash it.' He waited two whole weeks before he deposited it."

This might become a reality for more people in the future as well. Inequality widens when the rate at which wealth grows — say, your stocks or the price of your house — grows faster than the rate at which wages do. Research suggests that wealth is growing at a breakneck pace, keeping in line with economist Thomas Picketty's prediction of a dramatically inequal future.

Solutions for this will need to be found, and many municipalities or private individuals may find such a solution in constructing tiny homes. Homelessness is a powerful, self-perpetuating force, and having shelter is an obviously necessary step to escape poverty.

Regrettably, if tiny homes are being driven primarily by resource-conscious but fundamentally economically secure individuals, we can expect the trend to remain just that; a trend. In a few years, fewer and fewer tiny houses will be constructed and sold, and eventually there will just be a small contingent of diehard proponents of the lifestyle. If, however, the tiny home trend is being driven primarily by economic inequality, then we can expect it to stick around for a while.