Democratizing Data

with Erin Baumgartner

More Info

Show Notes

A look at how technology is democratizing access to products, services, and knowledge across the globe with entrepreneur and MIT researcher Erin Baumgartner.

About Erin Baumgartner

Erin Baumgartner CEO and Co-Founder Family Dinner Program Manager, MIT’s Institute for Data, Systems, and Society Former Assistant Director, MIT Senseable City Lab


TK: Welcome to Foresight Radio. I’m your host, Tom Koulopoulos. On each episode, we explore the many trends that are shaping the way we will work, live, and play in the future. Our focus is on disruptive and transformational trends that are changing the world in ways that are often invisible. Our objective is simple, give you the knowledge and the insights that you need to better manage the future. Foresight Radio is sponsored by our good friends at Wasabi. You can learn about them at In this episode we’re going to be talking about the democratization of data. As we enter a postindustrial era that will need to educate, feed, and care for 10 billion inhabitants. Our guest is Erin Baumgartner. Erin is the co-founder of an innovative new food deliver service, Family Dinner, which uses data in a novel way to bring local produce to your doorstep. She’s also currently working at the MIT Institute for Data Systems and Society on their MicroMasters program, a novel new open classroom that’s available to students around the world. Erin also is the past assistant director at the MIT Senseable City Lab, which looks at the data underlying the way our cities and urban society works. My conversation with Erin looked at a pretty broad range of topics from how data is being used to understand our urban landscape and reshape cities, to what’s using creating hyper personalized local food choices, and the way it’s helping to educate billions of new minds that would, otherwise, never have access to some of the world’s most elite schools. Here’s my conversation with Erin Baumgartner. Erin, we first met when you were at the MIT Senseable City Lab. You’re their assistant director. Tell us a bit about the Senseable City Lab. EB: The Senseable City Lab sits within the Department of Urban Studies and Planning at MIT. The goal of the lab is really to find ways to use data and data analytics and machine learning to make cities better from a variety of perspectives. We look at transportation, waste, environmental factors, and use data to really transform our understanding of those systems and try to find ways to make them better. Not better just so that they’re smarter and more efficient and green, which is great but also better for the people who live in them. The lab, at the end of the day, is very data focused but also very human focused, understanding that humans are at the nucleus of every city and to try and to make cities better for the people who live and work in them. TK: We hear an awful lot lately about urbanization and its tremendous flight to cities. EB: That’s right. TK: In many ways, it is over stressing our cities. How is that changing and making cities more pleasant to live in given the burgeoning growth? EB: I’ll point to a piece of work that the lab started several years ago around understanding taxis. The lab studied all the taxicab data for the City of New York for one year and that’s studying about 150 million rides to try to understand how many of them were redundant, like how many of them could be shared. We found that through two simple parameters, meaning people being willing to wait five minutes and willing to share the ride with one other person, that up to 90% of those rides are shareable. That’s just with the current system. That’s not changing to autonomous vehicles or anything like that. Ninety percent of those rides are shareable, so if you can reduce the number of taxis in the City of New York, the number of Ubers, the number of personal vehicles, how does that change the city? Noise pollution, air pollution, speed of traffic? What does it change for things like emergency vehicles, delivery vehicles, all of those things? Everything gets better across the board if you can find ways to optimize on the system. TK: That sounds awesome. Is anything being done as a result of that research? EB: Sure, yes. This research was published in a top tier academic journal. The lab started working with Uber on exactly this question, so Uber launched UberPOOL but how do you improve UberPOOL? How do you improve dispatching such that there aren’t cars that are driving around vacant? Now fast forward to autonomous vehicles and what does it look like when vehicles are autonomous and doing all this work instead of cars being parked 90%, 95% of the time? Do you need garages? Do you need parking lots? What does the city look like if it isn’t covered with parking garage? All that is very exciting for me. TK: Technologies like that make it so challenging to envision what the future will look like because we can’t really understand all the implications of what vessel of a smart city might feel like, how it will operate, but it’s all based on data. It all comes back to the data and the more data we have, the better we understand that data, the more decision we can make as a result of it to be more effective or more efficient. EB: That’s exactly right. Something the lab really excels in is data storytelling. All these data exists, what do you do with it? How do you ask it questions? Once you’ve run your analytics and all the smart people have run their algorithms and run their models on the data, how do you take what the data tells you and share it with a broader audience? The lab does a beautiful job of that, the data visualization, the data storytelling that the lab does in order to take very complex systems and turn them into something relatable is very important because it’s not just being able to take the data, study the data. It’s being able to make decisions on that data so make data driven decisions and also share it with the public, right? Autonomous vehicles are scary for a lot of people, but if you do a decent job of explaining the value of it and explaining that the technology is safe, then you can get people on board, but you have to do that storytelling first. TK: What was the coolest thing that you saw when you were at the Senseable City Lab? Was there one that’s an “aha” moment? You didn’t expect it but it really blew you away? EB: Yes. The lab started doing a project several years ago called Underworlds. In the Underworlds project, they were creating machines, semi-autonomous machines that sampled sewage, so robots in the sewage basically. TK: Sounds like a really bad Hollywood script. EB: Right, exactly. [Laughter] Or like a really bad graduate student experience, right? [Laughter] Like somebody had to build this thing and somebody had to test it, but they were deploying it in the city because we know that sewage is being tested but it’s being tested in wastewater treatment facilities like the one in Boston. By the time material gets there, it’s come from all over the city and so there’s no way of understanding, in a very geographically specific way, the health of the neighborhood. In the sewage, you can find all sorts of things - viruses, pathogens, drugs, whether they’re pharmaceutical drugs or illicit drugs. You can find all of that information in a hyperlocal way, in near real time by these machines that the lab built in-house. TK: Someone once said to me the world is a bunch of questions waiting to be asked and in many ways, this data raises the questions that we didn’t even know or didn’t think we could ask. EB: Sure. That’s exactly right. That’s something that the lab and academic environments really does beautifully because we’re not tethered to financial gain. You can just sit around and say, “What if we did this? What if we built robots and put them in the sewers in the City of Cambridge? Could we do that? Could that work? What could that tell us?” Then you go to another lab. We went to the Department of Biological Engineering to ask them, “Would this be feasible? Could this work?” Then the idea which was just a crazy seed on a whiteboard gives growth to something bigger, something much larger, and then you have to go to the City of Cambridge and say, “Hey listen, [Laughter] we’ve got this wacky idea. Would you pull up these manholes for us and let us do this?” Now, it spun off to a company called BioBot. TK: Here’s BioBot’s Newsha Ghaeli, president and co-founder, talking about how they use sewage data to report on urban health in real time. NG: Whether it’s opioid overdose up 200%, or disease spreading through our communities, or poisoning like lead in our drinking waters, now what all these headlines have in common is that we only hear about them when crisis turns to catastrophe, when many people have been affected and there’s nothing left to do but damage control. Now, it doesn’t have to be this way. Imagine if we knew the health of everyone in our cities in real time while people were still alive, before people got sick. Imagine a human health database powered by everyone in the city. At BioBot Analytics, we’re not only imagining this, but we’re building this. We measure human health in sewage. We tap into what you flush down the toilet everyday as it’s such a rich source of information on your health and wellbeing. EB: They’re doing this test around the world. It’s got real implications. The City of Boston, the City of Cambridge is really interested in using this technology to understand the opioid crisis. What we have on opioid understanding is deaths, its fatalities, but we don’t have really clear granular data on usage. Where are people using these? Where in the city? Not so that we can go down and track people down, but so that the federal dollars that are coming into the city, we know how to spend them and where a little bit better. Because without the data, people are just guessing. TK: The data also gives us a sense for, in many cases, how far we have to go. I think about autonomous vehicles and how we accept, without questioning it, much the fact that we kill over a million people a year globally with human driven vehicle. Yet someday, there’s no doubt we’ll look back on that and say how could we have lived that way? The same is true of the data that you’re uncovering, the Senseable City Lab is uncovering. We look at it and we ask the question, “Why do we put up with this?” [Laughter] There has to be a way to do it differently. EB: That ties into what we’re going to talk about later with the farming business and the industry in that way because I hope in the same way that we look back when we’ve improved on the system and say how could we possibly have allowed those things to happen. TK: I want to make that transition with you. I’m glad you went there because you’ve given us a perspective that you have from academia of what can be done and it’s fascinating. I’m sure the City of Cambridge is accustomed to getting very strange requests from MIT, right? EB: [Laughter] From MIT. TK: What a fascinating way to experience the world and that’s with academic laboratory environment, but you’ve made the transition. You have a startup, Family Dinner, and tell us a little bit about what you’re doing there because that is pretty far removed from the academic setting but in some ways, it has connected this notion of doing things differently and challenging the status quo. EB: You’re exactly right. Even though they seem rather disparate, I think they’re actually connected in many ways. What I learned from MIT and my work in the Senseable City Lab, and currently in my work at the Institute for Data, Systems, and Society, it really feeds into what we’re doing with this business, Family Dinner. Family Dinner is a farmers market delivery service. Each week, we gather fresh meat, produce, fish, cheeses, pasta, cookies and we aggregate it and we deliver it directly to people’s homes. We’re focusing on local farms, ethically sourced food, organically raised food because we really want to elevate what we think are the better practices in food. We also then send to people an email that says, “Okay, this is what you got. This is who grew it. This is where it came from. Here’s how you cook it. Here’s how to make sure that you’re not wasting any of it. Here’s what you do with your leftovers.” My husband’s been dying to write a cookbook called “Don’t Throw That Away, I’ll Eat it.” [Laughter] TK: I love that. EB: Because we’re just trying to focus on zero waste. TK: I’ve travelled to India. One of the things that amazed me in India was that you would actually see signs in cafeterias and restaurants asking you not to throw things away. Even at a buffet, take only what you can eat. We don’t have that ethos in the U.S. EB: Not at all. I think that this may be sound extreme but I find supermarkets, and airports, and buffets just extraordinarily depressing. It hurts because I walk through the supermarket and I have an understanding of how things were raised, where they came from, the impact of all that food, and I know that all of that work and all of that in some ways suffering, and all that environmental impact is, in so many cases, just going to end up in the trash. What we’re trying to do is push that aside and bring in a new model. TK: In the industrial era, the price of choice was waste in many ways. When you walk into a supermarket, a grocery store, and you had this enormous array of food that you could choose from, there was no doubt much of that would go to waste. That was the price of choice in the industrial era model. What you’re doing, however, is you’re taking much of the friction out of that model and I’m fascinated by much as the efficiency of it, but the experience now for the consumer is very different. In some ways, they have more choices and more choices that conform to their values, their way of wanting to experience the world. Talk a bit more about that. EB: We’ve got enough choice within our offerings that we can conform to different diets, different needs, allergies, and all that. Every week, I work with the farmers, and the fishermen, and the bakers, and the various providers to find out what’s new, what’s fresh, what do you have and try to think about what could go together and have the vision of what could this week share look like and then we bring it directly to people. I think the value for people is higher because it conforms exactly to what you’re saying, to their personal ethos of how they want to eat, how they want to spend their dollars. TK: A lot of this has to come back to the data. I’m sure you’re developing an understanding of individuals and your marketplace at a very low level, probably going back to a lot of what you experienced at MIT in the lab, using the data to better inform the choices and decisions. EB: That’s exactly right. We laughed that word, like the stitch fix for food. What we’re doing is curating to people’s taste, we’re moving inventory around, we’re using data science, software, logistics, and as much automation as we can to make this as optimized as we can. A lot of what I had understood about the need for data to inform decisions at MIT is coming to this business. I’ll give you an example, what we’re able to do is through the data science of collecting everything that we’ve got in from our customers and understanding where the growth is at, meet with farmers to say, “This is where we’re going to be in six months,” and why is that valuable to them because it helps them do their crop planning. This week I needed 300 pounds of potatoes, but understanding where the growth is going to be, next year you need to be planting me a thousand. The same with animals, if I’m going to need four cows, somebody needs to know that in advance so that we’re planning for it so that there’s no waste, and so that everything is being done in a timely manner and being done correctly. TK: In large part what you’re describing is a democratization of data because I would imagine that the family farmer, who goes to the family market on Saturdays and Sundays, isn’t necessarily accustomed to thinking of the world from a data point of view, but you’re introducing this to them. EB: That’s right. TK: Talk to me a bit about that. I love that notion, democratizing data. How are you doing that? What’s the effect of that long term? EB: Sure. The effect of that for us is what we’re doing is we’re building a decentralized network of small farms. The farmer that goes to the market is now connected in a decentralized way to 20, 30 other farmers that we’re working with, allowing them to benefit from our collection and use of the data to really just allow them to do what their core business is - be farmers. We’re trying to let farmers be farmers and let us take care of everything else. Because the data is there, we’re able to do that and not muddy their waters [Laughter] with it necessarily. I think another thing about the data around food has to do with the consumer with this massive flood of data where you can know everything about the tomato that sits in your kitchen. People mock it this precious thing. You and I, we’re talking about this Portlandia episode where this couple is sitting at dinner and it’s like, “Could you tell me the chicken’s name? Was it friendly with the other chickens?” Yes, that’s a joke. It’s precious, but it’s actually real. You can understand where did your tomato come from, what were the food miles, what’s the impact, and increasingly, people are interested in understanding what were the working conditions of people who are raising that food. Now imagine you go up the food chain. Now you’re talking about animals. What were these animals fed, what sort of antibiotics, what conditions were they kept under, what were the conditions of the people that were working in, working with those animals? It doesn’t matter where you come in on the spectrum of things that you care about, whether it’s pesticides, whether it’s CO2 emissions, whether it’s food miles. You can come in somewhere in the story of that tomato or that cow and find something that’s important to you. All that data is readily available and what we’re trying to do is tell the story of the better version. Not the tomato that comes from China, but that’s grown perfectly so that it’s exactly the size of a McDonald’s hamburger bun. TK: Is that true? EB: That’s true. [Laughter] You can modify a tomato so that it’s exactly circular to be exactly the size of a McDonald’s hamburger bun, and that’s the majority of - so China is the world’s largest producer of tomatoes and the US I think is third or fourth on that list. TK: Might I add, entirely tasteless tomatoes. EB: That’s exactly right. You know the funny thing about all of this is that, yes, we care about how the things are raised. We care about how animals are raised. We care about the CO2 emissions because the local stuff just tastes so much better. Like a summer tomato and we’re right in tomato season. There’s nothing like the smell of a tomato fresh out of your garden. It’s got nothing on the taste of the thing that you get from Mexico in January in the supermarket. TK: Let’s call a spade a spade. You see locally grown on so many products. The question is, where is local again? [Laughter] EB: That’s exactly right. These words [Laughter] - this is why I’m not allowed in the supermarket because it just greatly depresses me and I’m throwing these fits all over the place, but you see these words banded about constantly: local, fresh, nutritious, and everything’s got the word Earth and farm on it. Those words don’t actually have any metrics to determine what their meaning is, local to where exactly? People ask us about Blue Apron a lot. Blue Apron is one of those services that when you get this box from them, it’s got the word “local” written all over it, but they’ve got three distribution centers in the United States. One is in Jersey, one is Texas, and one is in California. I suppose like local to where exactly? When we say local, we mean that your food is grown in Massachusetts. We’re right over the border in New Hampshire. There’s a beautiful hydroponic lettuce facility in East Boston, so when I mean local, I mean your lettuce took the Blue Line to get to you. Those words are thrown about because people think they care about them. “This is local. This is nutritious. This is fresh.” What does that even mean? They know they care about it, but they haven’t actually understood and done a second level of looking into it to find out what the hell does that even mean. TK: For those folks outside of Boston, the Blue Line is one of our commuter railways here in Boston. Let’s talk about data again. It is quite possible that every vegetable, every fruit you buy, every piece of produce can be identified throughout its entire life cycle, from the seed that initially planted it, and I actually see a very sincere form of interest in that on the part of millennials and Gen Z. They want to know what that life cycle looks like, so this isn’t just a casual conversation. This is becoming a real market demand. People want to know this information. EB: That’s right. I hope it’s not just millennials. I hope that this message spreads. I was thinking about this the other day. What was it like - 15, 20 years ago, when people really started to think about where their clothing came from. There was this conversation around sweatshops and it started in the small, once again, little precious way, but it really grew into something much larger and had an effect. There was like a sea change farther down the road of how factories were treating their workers, minimum wages, maximum number of hours that people could work. This small conversation grew into something much larger and I think it’s time for that to happen with food as well. Yes, if you understand the whole story of your tomato on all the levels that I discussed, so the miles and the CO2 emissions, if you understand all of that, do you value it more? Do you value your banana even though it’s a little more brown than you normally would care for it and eat it instead of throw it away because you understand everything that went into it? That’s even more important when you’re talking about animals and understanding. If you look just a little bit deeper into the data to understand that factory farming produces 99% of the meat and meat products in this country. We’re feeding tens of millions of pounds of antibiotics in non-therapeutic ways to these animals. TK: You Are What You Eat Eats. EB: That’s right. That’s Michael Pollan, You Are What You Eat Eats, and it’s not just is it grass fed, is it corn fed? It’s the antibiotics or in the area of produce, it’s the pesticides. We should care about that because you’re consuming them and now you’re feeding them to your children. It’s not just the conversation that out in the ether because I’m just some Cambridge, Massachusetts hipster. I think it’s an essential thing to people’s health and it’s an exciting story to look into. TK: Erin mention that quote, You Are What You Eat Eats by Michael Pollan. Pollan is a popular writer with five New York Times best sellers. He focuses on agriculture and food. One of his better-known books is The Omnivore’s Dilemma. Here’s a short excerpt from that book. “Much of our food system depends on our not knowing much about it, beyond the price disclosed by the checkout scanner. Cheapness and ignorance are mutually reinforcing and it’s a short way from not knowing who’s at the other end of your food chain to not caring, to the carelessness of both producers and consumers that characterizes our economy today. Of course, the global economy couldn’t very well function without this wall of ignorance and the indifference that it breeds. This is why the American food industry and its international counterparts fight to keep their products from telling even the simplest stories - dolphin safe, humanely slaughtered, et cetera - about how they were produced. The more knowledge people have about the way their food is produced, the more likely it is that their values and not just value will inform their purchasing decisions.” That’s Michael Pollan from his book, The Omnivore’s Dilemma. Now back to our interview. What’s fascinating about the way you described that, Erin, is that the data connection - and we talked about technology a lot on this show, but this data connection actually gives - and this going to sound quite - it gives the tomato a narrative. It gives it a voice, this transparency out in the food [pride] self. I’m not taking your word for it. I can actually identify through the tomato where it’s been, how it was grown, how many resources were used to grow that little tomato. It sounds a bit absurd to us today, but only because we haven’t yet experienced that level of data connection at that low level of granularity, but we’re almost there. We’re getting there. EB: Well, that’s right, and we didn’t use to have it. We didn’t use to need this narrative because if you’ve been to Europe or you’ve been to Italy or Paris, and you’ve been to those outdoor markets, the person who’s feeding you those olives is the person who grew them 20 miles away. The person who’s giving you the bread had made it themselves that morning. The person who’s giving you the strawberries had grown them themselves, again, 20 miles away. That’s the way everybody used to shop. We didn’t need to understand the data because everything was very much hyperlocal. You didn’t really need the data because you had the interaction with the farmer directly in front of you who sold you what you needed. Now, everything is much more macro and so you need to look at the data. You need to look at where things are being grown in places that maybe don’t even meet the United States standards for food. What we’re trying to do is somehow go back to that smaller local relationship that we used to have. I think that this business allows us to do that. TK: Look, the other piece of this is that we are somehow going to have to change the way we feed the planet to be able to accommodate 10 billion people. The current models we have are not going to scale. Their inefficiencies, their wastefulness is not going to support 10 billion people. You’re still in MIT, however, and you’re at of all things to bring this full circle at the Institute for Data, Systems, and Society which is exactly what we were just talking about. EB: That’s exactly right. It feels like it was made to be. TK: What’s the connection there? What are they doing? What are you doing there? EB: The Institute for Data, Systems, and Society is one of the largest data science and statistics groups at MIT. It involves over 80 faculty members and they’re doing really extraordinary things. What I’m working on is the creation of a MicroMasters program in Statistics and Data Science that’s offered to anybody in the world. It’s got 23,000 people signed up for the first course. It’s made up of four courses so probability, statistics, data analysis, and machine learning with Python. The reason I’m so excited to be a part of this is its part of the democratization of education. MIT is an elite institution, but it’s got finite space. There are 4,000 undergraduates every year, but that doesn’t mean that there are only 4,000 super bright people. What this program is doing is allowing anybody in the world access to this very top tier education for free. If they want to receive a certificate, there’s a small fee, but it’s really nominal. TK: Why a master’s program? What was the reason behind that? What are the prerequisites to be able to get to the program? EB: There are no prerequisites. Anybody can take it. It’s important to know that it is MIT depth and rigor. This is not MIT Light. This is not something MIT is throwing out to the world to try to make money. It is bringing MIT quality education to the world. The reason that it’s a MicroMasters program is that there is no master’s in data science in the residential programs at MIT. We think that that may change over time, but we know that these topics, so probability, machine learning, data analytics are radically changing the world across sectors. The thing that’s interesting to us is putting it out into the world and allowing people to fix their own little corner of the world with that work, with that understanding, and with that education that maybe they didn’t have access to before. TK: There are two things that I want to explore you on that topic. The first is we often hear this new term, this new phrase that data is the new oil. In some ways, there’s truth to that but I think it over simplifies what data is. Data is much more complex than oil. Oil is pure commodity. Data manifest itself in many, many unique ways. However, it certainly is what’s filling the economy going forward. Data science analytics and understanding how to manipulate data and to understand it is key to all of that. On the one hand, we say, “Well, AI will do that for us.” When I look at data scientists today, much of data science is very manual. It’s very human intensive, not AI intensive. What do you see happening there? What’s the collaboration between the human being and the AI when it comes to data science? EB: Yes, data is everywhere. Yes, the cities and the entire landscape is blanketed with data in the digital exhaust. TK: Erin talks about the term digital exhaust, which is the byproduct of data that’s created by today’s digital devices. In my book, Revealing the Invisible, I describe the advent of the digital self and the digital twin, which are collections of data that intimately describe a person or an object in ways that make it possible to predict future behaviors. The amounts of data that are involved in the creation of these digital twins is staggering. Imagine that every person and object will soon have a digital counterpart that contains the entire history of behaviors throughout his, her, or its life cycle. For example, in the case of an Airbus 350, each wing alone has 10,000 sensors. In total, a typical A350 in service will generate nearly 10 terabytes of data each day. However, digital twins will not just be limited just to aviation. Every machine and nearly every object will have a similar collection of data. The volume of data created by these digital twins will eclipse anything we’ve experienced today as much as 400 exabytes. That’s roughly 300 times as much storage as exist in total today will be required just for the industrial data created by digital twins. EB: The work of data scientist is to gather that data, but also to ask really interesting questions of it that the data can’t ask itself questions. It’s for us to think about what are our needs, again, in our small corner of the world or what are the great needs of society and how can data feed answers, feed solutions into those pressing issues? I would say that AI is still very human focused and human-centric for now because of this need to ask questions of it, because it’s still so very new and I think that with the increasing number of data scientists that we’re trying to train exactly through programs like this, the questions that we ask of the data will become even more fine grained and even more specific to people’s localities and people’s need. TK: That last point is one that I wanted to continue just a bit because we sometimes develop this arrogance around data science and data analytics. The reality is that asking the questions in a very small geography can be critically important, knowing what that question is as opposed to what the broader question might be globally. Educating 22,000 people around the world serves that purpose too. You’re not training data scientists that will look at these enormous data collections that apply to global population, but folks that might be using small data collections that apply their local society, their village, their community. EB: I’ll return to the Underworlds that we mentioned. Whenever you start with the technology, you can’t even imagine all the implications it could have. All of the good it could do. We, for example, never thought about opioids and it hadn’t crossed our minds, but once you start having more and more conversations with people, they’re like, “Could this data tell us this? Could this machine do this for us?” We’re like, “Well, yes, it could,” and they’re like, “That would be enormously helpful.” We haven’t even thought of it because that’s not what’s at the forefront of our minds all the time. Once you start putting technologies in people’s specific context, they start thinking about how it could pivot to meet their needs and how it could pivot to make their lives better in that particular place. That’s essential because otherwise, if we’re not having those conversations and we’re not thinking about ways in which technologies can be adaptive, then what we turn into is that saying of, “I’ve got a hammer, everything looks like a nail.” TK: Furthering that, that’s what democratization again that we were talking about earlier, but in this case, from a standpoint of democratizing the tools used to finance to these very local specific questions. EB: Exactly. TK: You’re probably familiar with the film, The Graduate. It’s much older than you are, I know. However, there’s this famous line, “plastics,” right? [Laughter] What data science, in many ways, seems to be the career of the [consumer] pursuing these days. If data is the new oil, then it would make great sense? Do you agree? EB: Machine learning, artificial intelligence, data science is changing absolutely everything. I think the future generations of children will find themselves out in the cold if they don’t have a technical capacity, if they don’t have the ability to understand this type of work because it’s radically transforming every single sector. TK: It becomes a competence that anyone has to have, an understanding of it at least. EB: Absolutely. There will be very, very few things that won’t be touched and changed radically in the next 20 years by these tools and by these areas of academic study. TK: I have to ask you this as one of the topics because I often ask this question of guests in the program. What do you think about the apocalyptic view of AI? Is this something that we should be concerned about? If so, why? If not, why not? EB: I’m from New England. I’m pessimistic by nature [Laughter] so I see the parts of gloom and doom as far as reduction of the workforce. I don’t think that everybody who loses their job is going to be able to pivot and become the machine learning expert through retraining. I think that’s really almost dangerously Pollyannaish. I am very hyper aware of AI and machines interfering with people’s everyday lives through social media, in politics, all of that makes me very nervous, but at the end of the day, I’m extraordinarily optimistic because it will radically transform our lives for the better. Autonomous vehicles are a beautiful example as far as what they’re going to be able to do. Of course, emissions reduction, regaining green space, all of that I believe in, but back down to the idea of getting back to people, what could autonomous vehicles do to your 80-year old grandmother who can’t leave her house? Very sophisticated technologies could have very simple benefits so at the end of the day, I’m very optimistic. The thing I hear a lot in the gloom and doom factor is around privacy. Like, what about our privacy? I would always get that question at Senseable City, “Yes, but what about our privacy? Isn’t this Big Brother-ish?” I have a gloomy doomy answer to that one which is, “Your privacy ship sailed 15 years ago. It’s gone. You’re not getting it back.” If you have an email account, a bank account, a phone, you take public transportation, you go to the hospital, you live in a city, your data is everywhere. TK: Perhaps more to that point, I wouldn’t give up any of things that you just listed to get back some privacy because there are values in those. EB: You let Google know exactly where you are within a hyperlocal and a very granular way every time you use Google Maps. The Google Maps is this extraordinary tool. It’s extraordinarily valuable, and so of course, I give that up happily. The benefit that I get outweighs what they take from me in enormous ways, I think. TK: Can we talk about AVs a little bit more? EB: Sure. TK: I’m fascinated about this topic as it applies to your perspective because at the Senseable City Lab, I’m sure you talk about AVs a lot. One of the things that you often hear is that AVs will change the landscape of the city by getting rid of parking spaces, garages. What are some of the other things that you think will change because of AVs that we don’t necessarily see or talk about? EB: One of the things that we had looked at in the Senseable City Lab was autonomous intersections. One of the greatest causes of congestion in a city and emissions in a city are traffic lights. Traffic lights are the arbiters of space and flow of traffic, but they’re incredibly inefficient and many of them aren’t smart in any way. Never mind autonomous. When vehicles become fully autonomous, will you need the intersections anymore? Will you need red lights and green lights? Now, red lights/stoplights are technology that is 150 years old. It was originally designated for horses. In that 150 years since its genesis, it really hasn’t changed a whole lot. Now you still have to wait until the red light changes and then you go. Will you need those red lights and green lights anymore when cars are fully autonomous? What the lab did - TK: Anyone whose driven, by the way, in India knows that [Laughter] it’s purely optional. EB: I get that joke all the time. What the lab did was create a model for a fully autonomous intersection called a slot-based intersection model. It borrows its ideas from air traffic control. Your fully autonomous car will arrive and traverse the intersection only at a time and space during which it is safe to do so. It increases the flow of traffic by something like 40%. It decreases emissions by something like 60%, but now you’ve got a whole new set of problems, where do the people go? If it’s a brand new city, sure, you can build under or above an intersection, but on a legacy city like Boston, what are you going to do? The art of what’s possible has to meet the everyday need of what’s there and that’s sometimes where the friction is, but you could turn that around and say, “That’s where the creative moments are.” TK: The landscape of the city of tomorrow might look radically different than that of today? EB: Right, but do you always want it to? Do you want Paris to change? There are things that I would change about Paris like lines, and the post office, and things like that but there are things I would never want to change. Walking along the sand at night is an extraordinarily beautiful inspiring thing. Would it feel the same if there are autonomous cars zooming around? I think that the landscape of cities will change, but I hope that the landscape of all cities doesn’t. TK: Let’s come back to this theme of democratization because you said a few things over the course of our conversation that I want to bring together. One was you talked about the MicroMasters program at MIT, then we talked about the democratization of the food industry and the way we’re seeing data democratization change and create new business models. What fascinates me is that a lot of times, democratization is a huge threat. For example at MIT, when you blow the doors open and say anyone in the world can now be a student effectively at MIT, are you, in many ways, potentially eroding or destroying the brand, and the stature, and the history of MIT as a very exclusive, elite organization? Talk to us about that. EB: I think MIT will remain one the most elite educational institutions in the world no matter what we’re doing with our online offerings. I think what we’re trying to do with the online offerings is bring MIT level education out to everybody in the world, but nothing can touch the residential experience of being at MIT for four years, surrounded by the brilliant faculty and the other brilliant students. Though we’re opening the doors and blowing the doors off of the model, I think the core of MIT will remain and that nothing can touch that. In addition, I think that by blowing the doors off, we’re also enriching that residential experience. There’s this story of when MIT started first putting its courses out through MITx online, one of the first courses in the computer science department received something like 150,000 people around the world, wanting to sign up and take it. That’s a greater number of people than ever received undergraduate diplomas from MIT in its 150 years. TK: When you say around the world, every region, every geography…? EB: Everywhere. A hundred and fifty thousand people signed up to take that class. Some number of people finished the course, some smaller number of people passed and some very small number of people received perfect scores. One of the people who received a perfect score was a 15-year-old kid in Mongolia who had been taking these classes online at his town library. TK: Not even at home, at the town library. EB: At the town library. There’s a beautiful article in the New York Times about it. MIT was able to look at the data on this offering and now bring this student to MIT. It’s wonderful. Through the online MicroMasters, we had a webinar yesterday and all these questions came in from around the world. One of them was from a kid in Turkey and he said, “I’m 14. Can I sign up for the MicroMasters in Statistics and Data Science?” I was like, “Heck, yes, kid! [Laughter] Of course you can. Do you have multi variable calculus in Python? Because you’ll need them to succeed, but of course you can,” and so it really reaches what MIT is trying to do because there’s no way that we know out of the 7.2, 7.3 billion people that are on Earth, there’s no way MIT has access to all of them. What we’re trying to do with the online offering is bring this top tier education to the world and it’s an incredible mission and I’m very honored to be part of it. TK: Nothing is more profound than what you just described because perhaps the least well leveraged and under utilized resource globally is human capital. We all have the same amount of gray matters. Certainly, some of us utilize it differently than others but so much of it never has the opportunity to ever come close to realizing its full potential. In part, what you’re doing is creating a pathway through which at least some of those people can. All of this is enabled by the democratization of data education in this case, but it’s data. What happened to the Mongolian student at MIT? Did he…? EB: I don’t know the end of the story. I wish I did. I don’t know the end of the story but I know it’s not the only one like it. TK: As it turns out, the student that Erin was referring to, Battushig Myanganbayar, went on to receive his Bachelor’s in Electrical Engineering and Computer Science from MIT and he’s currently a research assistant in the MIT media lab. EB: I think that that’s really beautiful. If there’s one story like that, it makes the whole thing valuable. It makes the whole thing worth it and proof that in the face of gloom and doom of technology and access in blowing the doors off, there’s these beautiful little gems at the center of it that make everything worth it. TK: We’re very good seeing the fear first and then realizing the opportunity much later. That usually is the way progress works. EB: That’s right. TK: In closing, you said something to me earlier that I want to make sure we talk about just a little bit. Data is becoming much more than just the new oil. It’s becoming a new social participant. It’s becoming part of our conversation. It’s informing us in ways that we never had the opportunity to be informed before. Talk a bit about that. EB: I’ve been thinking a lot about what data does for us and I think that a way that I like to describe it is that data is this invisible journalist for us, an invisible investigator. It uncovers things that we never thought to think about. It uncovers truths that we didn’t know to ask about. All of that makes us more informed and changes the way that we shape our values, changes the way we allocate our resources, whether it’s our time, whether it’s our money. It’s really giving this whole new set of information to us that allow us to weave our lives in different ways. Again, it’s doing this in this invisible, behind the scenes way but it’s radically changing the way we do everything, whether it’s telling the story about your tomato or telling a story of a 15-year-old Mongolian boy taking computer science classes. The data is allowing us to see things that weren’t invisible, we just didn’t have an eye opened to them. TK: We didn’t even know how to ask the questions. EB: Right. TK: Data has a fundamental dynamic on how we shape society, commerce, the world. What a fascinating conversation, Erin. Thank you so much for joining us in Foresight Radio. EB: This was a joy. Thank you so much. TK: Good luck with the start-up, Family Dinner. is the website. We have a link to that on Foresight Radio. Thanks again, Erin. EB: Thank you so much. TK: That was Erin Baumgartner speaking with me about the democratization of data. To find out more about Erin, her company,, or the MIT MicroMasters, just click the link on the Foresight Radio homepage at Thanks again to our sponsors for this episode of Foresight Radio, Wasabi Technologies. Take a look at how Wasabi is changing the rules of the game when it comes to cloud storage at This is Tom Koulopoulos. I look forward to joining you again soon for another episode of Foresight Radio where we look at the future of how we will live, work, and play.