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 is the Peter Munk Professor of Entrepreneurship at the Rotman School of Management, University of Toronto. He is also a Research Associate at the National Bureau of Economic Research in Cambridge, MA. He is founder of Creative Destruction Lab, co-founder of The Next 36 now known as NEXT Canada, and Academic Director of the Centre for Innovation and Entrepreneurship. Agrawal is co-founder of an annual conference, held at the University of Toronto, on the business of artificial intelligence, “Machine Learning and the Market for Intelligence.” Agrawal is a co-author of the book Prediction Machines: The Simple Economics of Artificial Intelligence and co-editor of the forthcoming book "The Economics of Artificial Intelligence: An Agenda" (University of Chicago Press).
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.
- Push Past Negative Self-Talk: Give Yourself the Proper Fuel to Attack the World, with David Goggins, Former NAVY SealIf you've ever spent 5 minutes trying to meditate, you know something most people don't realize: that our minds are filled, much of the time, with negative nonsense. Messaging from TV, from the news, from advertising, and from difficult daily interactions pulls us mentally in every direction, insisting that we focus on or worry about this or that. To start from a place of strength and stability, you need to quiet your mind and gain control. For former NAVY Seal David Goggins, this begins with recognizing all the negative self-messaging and committing to quieting the mind. It continues with replacing the negative thoughts with positive ones.
Is this proof of a dramatic shift?
- Map details dramatic shift from CNN to Fox News over 10-year period
- Does it show the triumph of "fake news" — or, rather, its defeat?
- A closer look at the map's legend allows for more complex analyses
Dramatic and misleading
Image: Reddit / SICResearch
The situation today: CNN pushed back to the edges of the country.
Over the course of no more than a decade, America has radically switched favorites when it comes to cable news networks. As this sequence of maps showing TMAs (Television Market Areas) suggests, CNN is out, Fox News is in.
The maps are certainly dramatic, but also a bit misleading. They nevertheless provide some insight into the state of journalism and the public's attitudes toward the press in the US.
Let's zoom in:
- It's 2008, on the eve of the Obama Era. CNN (blue) dominates the cable news landscape across America. Fox News (red) is an upstart (°1996) with a few regional bastions in the South.
- By 2010, Fox News has broken out of its southern heartland, colonizing markets in the Midwest and the Northwest — and even northern Maine and southern Alaska.
- Two years later, Fox News has lost those two outliers, but has filled up in the middle: it now boasts two large, contiguous blocks in the southeast and northwest, almost touching.
- In 2014, Fox News seems past its prime. The northwestern block has shrunk, the southeastern one has fragmented.
- Energised by Trump's 2016 presidential campaign, Fox News is back with a vengeance. Not only have Maine and Alaska gone from entirely blue to entirely red, so has most of the rest of the U.S. Fox News has plugged the Nebraska Gap: it's no longer possible to walk from coast to coast across CNN territory.
- By 2018, the fortunes from a decade earlier have almost reversed. Fox News rules the roost. CNN clings on to the Pacific Coast, New Mexico, Minnesota and parts of the Northeast — plus a smattering of metropolitan areas in the South and Midwest.
Image source: Reddit / SICResearch
This sequence of maps, showing America turning from blue to red, elicited strong reactions on the Reddit forum where it was published last week. For some, the takeover by Fox News illustrates the demise of all that's good and fair about news journalism. Among the comments?
- "The end is near."
- "The idiocracy grows."
- "(It's) like a spreading disease."
- "One of the more frightening maps I've seen."
- "LOL that's what happens when you're fake news!"
- "CNN went down the toilet on quality."
- "A Minecraft YouTuber could beat CNN's numbers."
- "CNN has become more like a high-school production of a news show."
Not a few find fault with both channels, even if not always to the same degree:
- "That anybody considers either of those networks good news sources is troubling."
- "Both leave you understanding less rather than more."
- "This is what happens when you spout bullsh-- for two years straight. People find an alternative — even if it's just different bullsh--."
- "CNN is sh-- but it's nowhere close to the outright bullsh-- and baseless propaganda Fox News spews."
"Old people learning to Google"
Image: Google Trends
CNN vs. Fox News search terms (200!-2018)
But what do the maps actually show? Created by SICResearch, they do show a huge evolution, but not of both cable news networks' audience size (i.e. Nielsen ratings). The dramatic shift is one in Google search trends. In other words, it shows how often people type in "CNN" or "Fox News" when surfing the web. And that does not necessarily reflect the relative popularity of both networks. As some commenters suggest:
- "I can't remember the last time that I've searched for a news channel on Google. Is it really that difficult for people to type 'cnn.com'?"
- "More than anything else, these maps show smart phone proliferation (among older people) more than anything else."
- "This is a map of how old people and rural areas have learned to use Google in the last decade."
- "This is basically a map of people who don't understand how the internet works, and it's no surprise that it leans conservative."
A visual image as strong as this map sequence looks designed to elicit a vehement response — and its lack of context offers viewers little new information to challenge their preconceptions. Like the news itself, cartography pretends to be objective, but always has an agenda of its own, even if just by the selection of its topics.
The trick is not to despair of maps (or news) but to get a good sense of the parameters that are in play. And, as is often the case (with both maps and news), what's left out is at least as significant as what's actually shown.
One important point: while Fox News is the sole major purveyor of news and opinion with a conservative/right-wing slant, CNN has more competition in the center/left part of the spectrum, notably from MSNBC.
Another: the average age of cable news viewers — whether they watch CNN or Fox News — is in the mid-60s. As a result of a shift in generational habits, TV viewing is down across the board. Younger people are more comfortable with a "cafeteria" approach to their news menu, selecting alternative and online sources for their information.
It should also be noted, however, that Fox News, according to Harvard's Nieman Lab, dominates Facebook when it comes to engagement among news outlets.
CNN, Fox and MSNBC
Image: Google Trends
CNN vs. Fox (without the 'News'; may include searches for actual foxes). See MSNBC (in yellow) for comparison
For the record, here are the Nielsen ratings for average daily viewer total for the three main cable news networks, for 2018 (compared to 2017):
- Fox News: 1,425,000 (-5%)
- MSNBC: 994,000 (+12%)
- CNN: 706,000 (-9%)
And according to this recent overview, the top 50 of the most popular websites in the U.S. includes cnn.com in 28th place, and foxnews.com in... 27th place.The top 5, in descending order, consists of google.com, youtube.com, facebook.com, amazon.com and yahoo.com — the latter being the highest-placed website in the News and Media category.
- Master Execution: How to Get from Point A to Point B in 7 Steps, with Rob Roy, Retired Navy SEALUsing the principles of SEAL training to forge better bosses, former Navy SEAL and founder of the Leadership Under Fire series Rob Roy, a self-described "Hammer", makes people's lives miserable in the hopes of teaching them how to be a tougher—and better—manager. "We offer something that you are not going to get from reading a book," says Roy. "Real leaders inspire, guide and give hope."Anybody can make a decision when everything is in their favor, but what happens in turbulent times? Roy teaches leaders, through intense experiences, that they can walk into any situation and come out ahead. In this lesson, he outlines seven SEAL-tested steps for executing any plan—even under extreme conditions or crisis situations.
SMARTER FASTER trademarks owned by The Big Think, Inc. All rights reserved.