Here is the transcript:

As-salāmu ʿalaykum. Good morning! Thank you all for coming out this morning. I’m very grateful for The Asian Banker for inviting me to share with you a few thoughts about engineering smarter cities and fintech. So, why are we talking about smart cities? I thought this was a Future of Finance conference? Well, more than 50% of the world’s population lives in a city today, and by 2050 about two-thirds of the world’s population is going to live in a city. You cannot talk about the Future of Finance without talking about the environment in which people are transacting in cities, so we’re going to talk today about collaboration and data analytics. Which we think we need both in order to carve this new future.

So, big idea number one. We’re going to throw a few ideas today. There’s an expression that MIT’s president of the 1970s Jerry Wiesner, who then helped found the Media Lab, said, “At MIT, education is like a drink from a fire hose.” And the students responded by taking a fire hydrant, chaining it to a water foundation, and bedding it in concrete. That was one of the more famous hacks on campus, but I’m going to give a little bit of MIT fire hose today. So we’re going to go fast, cover a lot of ideas, and hopefully leave you with some inspiration.

Big idea number one - how ideas flow through city predicts GDP. We looked at over 200 cities in North America and Europe, and this holds true that I know the rate of idea flow of a city, the pattern of phone calls, this aggregate data analysis, I can tell you the GDP trajectory of that city or neighbourhood. There’s only one city out of the 300 cities we analysed, where there’s a high rate of idea flow and a low rate of economic productivity. What’s the one city? Shout it out. What do you think is the one city where there’s a high rate of idea flow and a low rate of economic productivity?

I know some of my former students are here in the room, and I’m going to start picking on you if you don’t draw out some answers here. Anyone? Muhammad, take a guess. What’s one city with a lot of ideas but not much economic productivity? Boston? No, curiously, Boston should be the answer, but it’s not. Boston actually does have a lot of start-ups and established business. It’s Washington, D.C.

So big idea number two: there are 1.5 billion people in the world who lack a legal identity. Identity is the onramp for financial inclusion. How can we offer banking services, mobile banking, and payments if they don’t know who they are? Identity today is deeply flawed. The way we get identity is our parents give birth to us, they tell some doctors yes, yes, they’re really our parents, and the doctors talk to the government. In my case the US government issues me some paper that says I’m me. This model is very flawed. It’s centralised. It’s easily forged, and the cost of compliance is astronomical around that, so we all here probably deal with AML and KYC issues. I remember reading recently that Citigroup had a very successful cost cutting initiative. They cut over $3 billion through layoffs, automation, and all sorts of stuff, and they spent all of it on compliance, mostly dealing with AML and KYC. Our identity systems are not keeping up with the world we’re living in, but that mobile phone you carry around with you is actually a really powerful tool for robust identity, for cyber-secure identity. So we’ll talk about that a little more.

Something that we spend a lot of time thinking about at MIT is an operating system for smarter cities and a smarter financial system. I’m going to take you through what that means. This is a picture from our research group in 1996. It probably looks like what you think of when you think of MIT students. So what they were trying to do here. Back then I remember in the early 1990s, computers were about the size of the speaker, and even laptops were the size of a substantial briefcase, but the trend was getting smaller and smaller. There was a vision that someday, we would have these wearable devices and little computer that you could in the palm of your hand that would tell us more about the world around us. We wanted to simulate, so you had to actually strap a motorcycle battery on your waist to power electronics. They put on a mining helmet with a video camera, and then they tried to have the computer tell them what was going on around them. They called themselves the Cyborg Collective. In fact the guy there went on to invent Google Glass.

In 1995, we were visualising what the future would like, what these little, wearable devices that now everyone’s got, a smartphone. How many of you have an Apple watch or a smart watch? These wearables are starting to come into our lives, and this led to a new discipline called “social physics”, because once we had these devices on people, and we began to be able to measure how they were moving a city and how they were interacting with each other, we found that there were predictive equations and math that could describe what was going, what people were going to do, and how to influence it. This is really big stuff, because before that in the science of understanding, people was sort of relegated to the soft sciences. We called it sociology, and it really didn’t provide anything useful, so people made a lot of guesses. There’s a famous expression in the advertising industry, for example, “I raised 50% of my advertising spending. I just don’t know which 50%,” because the understanding of people was so poor.

Now there’s new math, social physics, which can tell you what people are doing, what they’re going to do, and how that interacts in a society. Forbes called Stanley Penley, an inventor and science professor at MIT, one of the seven most powerful data scientists on the planet, because of this discovery, this social physics. So, the curious thing is the ideas behind it were a couple of a hundred years old. How many of you studied economics like Adam Smith in school? You heard about the invisible hand, this idea that each of us individually are selfish, that we act in our own self-interest, but somehow when we interact in a market together, we miraculously sort of move in a direction that creates a stable market environment.

Now people forget that Adam Smith actually was not talking about selfish self-interest. He also said that people are doing all these activities that they’re doing for collective good. That they’re engaging in trade for collective benefit. That there’s this concept of collective intelligence. Now the problem was back in the 1700s he didn’t have the math. He didn’t have the computers to be able to really prove that out, but now we do. In fact we figured out how to calculate the incentives necessary to get a group of people to do stuff for societal benefit. So, what this leads us to is the ability to create an operating system for smart cities, for smarter societies, and for a smarter financial system. I’m going to talk about that in detail in a minute. Right now I’m just going to briefly take you through the principles of this operating system.

So we need a robust identity, because that’s how we get people into the system. We need distributed trust. If we had a robust identity, and we had distributed trust authorities, we could lower the cost of AML/KYC by three orders of magnitude. Think about that for a minute. Think how much your banks and non-bank financing companies spend on AML/KYC. By using open algorithms we can now unlock the power of analytics and make data available for sharing and for societal good without invading privacy and releasing personal information that people want to keep protect. I’m not going to go into distributed complication. Now I’ll just say there’s cool math and neat computer programs that can make this all possible.

And it’s really important that we include everybody. The risk of all this great technology is that we’re going to leave even more people behind. Universal access is important, so we have Accenture, Intuit, Microsoft, the World Economic Forum, the United Nations, the French government, and several others working with us on building this new operating system that we call the Trust Data operating system. There’s more on this. I’m not going to spend a lot of time on this now, but if you go to trust.mit.edu, there’s more information about this.

So what can we do with this? So now we’re getting to the exciting part, because we’re not just fascinated by technology. Our mission at MIT is to solve the world’s greatest problems through technology. It’s the impacts that we care about. In the region we’ve been looking at the questions of youth unemployment and can we use data analytics and the science of idea flow to improve employment. We can improve transportation, which in turn can improve GDP. Some of you have been able to work with a city of about four million by looking at information flows. By looking Telecom’s data, we helped improve everyone’s commute time by 10%.

We can manage scarce resources. There’s a government that wanted to reduce electricity utilisation, because on a certain level they could go all hydroelectric. They could also run on diesel generators. So by using social incentives, there was a little dancing bear that would appear on your electricity bill every month when you went to pay it, they were able to get everybody to use 17% less electricity. The way they did that is they may have not incentivised you with those dancing bears. They incentivised your neighbour. So, if you used less electricity, he would little marmots dancing on his screen at the end of the month. It seems silly, but it actually works. People care about what other people think, and it motivates behaviour. So, of course, why not water? Any resource could be managed with these methodologies.

We’re reengineering the economy of Andorra, which is this fascinating little country in the Pyrenees between France and Spain. It’s got 80,000 residents and 12 million visitors annually, so their entire economy is driven by visitors. You use data analytics to make that experience much more pleasant. And so now we go from smarter cities to smart fintech. We do our experiments on creating these smarter, data-driven environments, both virtually in digital trading systems, as well as physical in urban environments. So for example, we took a day trading network of 500 currency traders, and by optimising how they got trading ideas, came up with ideas with how to trade currency derivatives, we were able to improve everybody’s hourly in the entire market by 50%.

I teach MIT’s fintech class online, and we’ve gone into 130 countries. We took this global audience of students, and we were able to use this group to predict the closing price of the stock market. Prediction markets as an idea have been around for a while. Ten years ago people got excited about them. They said, “Oh, the internet is connecting everybody. Now we can predict things.” It turns out if you just take the averages, it doesn’t work very well. It’s plus or minus 5%, which is maybe great for sort of a parlour trick, but it doesn’t actually let you make intelligent trading decisions. So, if you just do the wisdom of the crowd, you’ve got a bit of an error mark around your prediction. If you just take the experts, people who are really good at making predictions, you basically get them disagreeing with each other because part of what makes an expert an expert is that they stake out an opinion, and over time they get pushed to stake out more and more extreme opinions. Nate Silver wrote a wonderful book called The Signal of the Noise, and he talks about this factor in retail. What we found is that if you tune the market, if you put a little more weight on the experts but not a lot, you can predict the closing price of the S&P 500 within .1%. So this is unpublished data, and but we are going to putting out a paper on this in the near future.

You can also use these predictive sciences, the social physics to come up with a better credit score. Let’s think about this. The models of credit risk that we manage our institutions with are 50, 60 years old. We’re using linear progression model. That’s what a credit bureau does. They gather all your credit history. They put it into a big database, and then you extrapolate forward and you say, “Past results are a prediction of future results.” Anyone who lived through the 2008 financial crisis knows that’s not true. So we used mobility, communications, and other behavioural data to build a better credit score. In fact 30% to 50% better than what banks are using today. And when I build a better credit, I don’t mean this was in some lab sitting a computer. I mean we worked with a real bank in predominantly Muslim country, that shall remain nameless, and we were able to figure out how people move around a city and how that related to their financial behaviour. That we actually spun out into a company called Distilled Analytics.

Now let’s get to the what ifs. I’ve just taken you through a bunch of different ideas about how we can use collaboration and data analytics to engineer better cities. What does that look life for the region? So what if we had a laboratory for a new fintech? What if we could take the city and turn it into this vibrant experimentation environment where you could try out new ideas? So that would require the regulator, industry, academia, innovators, all working together in this living lab for fintech.

What if Dubai unlocked Africa? Here’s an interesting fact. There are about a billion mobile communications lines now in Africa. There are over two hundred million smart phones already deployed in the continent, and there are going to be about five hundred million smart phones in Africa with three to five years. That’s going to change everything. Dubai and the UAE are in the unique position to be an enabler of that economic development. Africa is the last frontier market. It is a huge opportunity to drive economic prosperity and progress. The UAE has a very interesting position to help facilitate that.

What if Dubai connected a constellation of fintech cities? What if the location here is able to serve as a bridge between all great stuff going on in Singapore, and Hong Kong, and Vietnam, the Asian region of Shanghai with the activity going on in London, and Amsterdam, and New York, in Mexico City, in Sao Paulo? There are some tremendously exciting opportunities, and I am thrilled to be alive at this time.

There are more of our thoughts on things like fintech and trusted data at visionaryfuture.com. We’ve written some books on this, and I’m really excited to be part of the panel discussion now. Thank you for taking the time.