Skip to Content
MIT Technology Review

Visionaries (2014)

People who are reimagining how technology might solve perennial human problems.
  • Manu Prakash

    Age:
    34

    Imaginative inventions liberate science from the ivory tower.

    Manu Prakash is determined to push down the cost of doing science. Expensive facilities, he says, limit knowledge and expertise to a privileged elite. So from his lab in Stanford’s bioengineering department, he’s producing instruments that enable people to undertake scientific explorations on the cheap.

    Prakash’s ­Foldoscope is assembled like a paper toy.

    Many of Prakash’s inventions have a surreal quality. Consider his $5 microfluidic chemistry lab. At a holiday gift exchange, his wife received a hand-cranked music box that used a piano-roll-style punch tape to sound notes. Prakash recognized the mechanism’s potential to combine chemical reagents according to a program (the punch tape), without electricity (thanks to the hand crank), at a fraction of the usual cost. He now makes the tiny labs from scratch.

    Prakash was raised in northern India and has done fieldwork in Uganda, Ghana, and other developing countries, giving him a view of problems that might not be apparent in most well-equipped academic labs. His insights have led to devices like the Foldoscope, a research-grade microscope made of plastic-impregnated paper, which costs a mere 55 cents, and the OScan, a 3-D-printed smartphone add-on that helps diagnose the oral carcinomas that are responsible for 40 percent of cancer-related deaths in India. His aim, he says, is to put scientific tools in the hands of anyone with a question.

    Ted Greenwald

    Watch this Innovator at EmTech 2014
    Meet the Innovators Under 35

  • Rumi Chunara

    Age:
    32

    Crucial information about disease outbreaks can be gleaned earlier.

    Problem:
    Our systems for detecting outbreaks of disease are unreliable. Typically, word of outbreaks bubbles up as patients see health professionals, who report cases to authorities. Those authorities often can’t piece the reports together in time to prevent significant numbers of other people from getting sick.

    Solution: 
    Rumi Chunara, a researcher at Boston Children’s Hospital and Harvard Medical School, is mining social media and other online sources for information outside of medical settings. 

    In one study, Chunara found that a rise in cholera-related Twitter posts in Haiti correlated with an outbreak of the disease. “That’s important, because it takes the ministry of health in Haiti a couple of weeks to get their data aggregated,” she says. In future outbreaks, tweets could help direct medical workers earlier and ensure that supplies like water purification tablets get where they’re needed. 

    Chunara knows that digital disease detection is not necessarily better. For instance, researchers have spotted inaccuracies in Google’s Flu Trends service, which analyzes Web searches and estimates the pervasiveness of the flu. But her goal is not to supplant the traditional chain of command in public health; instead she is augmenting it with new tidbits of information.

    To get beyond what might be found from social media, she offered two-cent rewards to people in India who completed a survey about malaria, generating information that could guide deployment of diagnostic and treatment kits. For the United States, she helped develop Flu Near You, a site that creates flu maps based on user-submitted information about symptoms and diagnoses. She’s showing that you can get good data even from people who haven’t seen doctors.

    —Courtney Humphries 

    Watch this Innovator at EmTech 2014
    Meet the Innovators Under 35

  • Severin Hacker

    Age:
    30

    A novel approach to learning languages is making the Web more accessible.

    In 2009, Severin Hacker and Luis Von Ahn were holed up in the computer science department at Carnegie Mellon University, turning over a seemingly impossible challenge: how to translate all one trillion pages on the Web, which are mainly in English, for people who speak other languages.

    Neither Hacker, a native of Switzerland who was then a PhD student, nor Von Ahn, who was raised in Guatemala and served as Hacker’s advisor, was very impressed with the options. Feeding Web pages into Google Translate usually generated gobbledygook, while trying to hire enough translators was impossible. “Translating is a task that people don’t want to do,” Hacker says. “It is work.”

    So Hacker made a game out of it. Known as Duolingo, it teaches foreign languages to anyone with a smartphone or an Internet connection, for free. Unlike most language classes, with their reliance on rote memorization, Duolingo offers constant interaction. You respond to multiple-choice questions and complete sentences by typing in answers, and you practice phrases by speaking into the microphone. If you answer incorrectly, the app shows you where you went wrong; if you make too many mistakes in a section, you’ll have to repeat it. Each course takes around 35 hours to complete and promises intermediate-level proficiency.

    But the real genius of Duolingo is the way it solves the problem that first stumped Hacker and Von Ahn. When you reach the highest levels of a course and translate sentences into the language you’re learning, Duolingo compiles your work with that of other students. That aggregated work tends to produce accurate translations, and media companies such as Buzzfeed and CNN are paying Duolingo—now spun out as a company, with Hacker as CTO—for foreign translations of their English Web pages.

    Duolingo offers courses in 30 languages and counts 30 million users. Hacker himself recently used it to learn Spanish in order to travel to Guatemala for Von Ahn’s wedding. Relying just on the app’s lessons, he navigated the airport, hotel, and restaurants, read the newspaper, and got a haircut. “Ten minutes of Duolingo,” he says, “is worth, like, an hour of class.” 

    —Patrick Doyle

    Watch this Innovator at EmTech 2014
    Meet the Innovators Under 35

  • Sarah Kearney

    Age:
    29

    A financial innovator is crafting a way for foundations to invest in clean energy.

    Sarah Kearney wants philanthropists to act more like venture capitalists.

    She’s created a nonprofit called Prime Coalition to help private foundations invest in energy-related startups. Given that venture funding for clean energy technologies has dropped substantially in the past few years, Kearney’s idea could open up a source of much-needed capital from long-term backers.

    Radically new energy technologies have proved too risky for most private—and even government—investors. But foundations might not mind those risks, and they need to give away a small portion of their endowments every year if they are to maintain their tax-exempt status. Why not use it to fund companies whose primary goal is fighting climate change?

    So far, seven foundations have bought into her vision, giving her money to fund Prime Coalition. Next, she hopes to facilitate deals between philanthropists and startups working in renewable energy, energy storage, and related technologies.

    Creating the necessary legal and financial structures is a long, complex process. There are potential pitfalls, too. If a startup gets money from a chari­table foundation and private investors, will the foundation’s social mission be at odds with the goals of the profit-­seeking capitalists? But Kearney says foundations that want to do something about climate change should be involved with technology startups somehow. “Nothing is going to happen at scale unless people are making money,” she says.

    Martin LaMonica

    Watch this Innovator at EmTech 2014
    Meet the Innovators Under 35

  • Quoc Le

    Age:
    32

    Frustration with waiting for computers to learn things inspired a better approach.

    Growing up in rural Vietnam, Quoc Le didn’t have electricity at home. But he lived near a library, where he read compulsively about great inventions and dreamed of adding to the list. He decided around age 14 that humanity would be helped most by a machine smart enough to be an inventor in its own right—an idea that remains only a dream. But it set Le on a path toward pioneering an approach to artificial intelligence that could let software understand the world more the way humans do.

    That technology sprang from the frustration Le felt at the Australian National University and then as a PhD candidate at Stanford as he learned about the state of machine intelligence. So-called machine learning software often needed a lot of assistance from humans. People had to annotate data—for example, by labeling photos with and without faces—before software could learn from it. Then they had to tell the software what features in the data it should pay attention to, such as the shapes characteristic of noses. That kind of painstaking work didn’t appeal to Le. Although personable with other humans, he is uncompromising in his expectations for machines. “I’m a guy without a lot of patience,” he says with a laugh.

    While at Stanford, Le worked out a strategy that would let software learn things itself. Academics had begun to report promising but very slow results with a method known as deep learning, which uses networks of simulated neurons. Le saw how to speed it up significantly—by building simulated neural networks 100 times larger that could process thousands of times more data. It was an approach practical enough to attract the attention of Google, which hired him to test it under the guidance of the AI researcher Andrew Ng (see “A Chinese Internet Giant Starts to Dream”).

    When Ng’s results became public in 2012, they sparked a race at Facebook, Microsoft, and other companies to invest in deep-learning research. Without any human guidance, his system had learned how to detect cats, people, and over 3,000 other objects just by ingesting 10 million images from YouTube videos. It proved that machines could learn without labored assistance from humans, and reach new levels of accuracy to boot.

    The technique is now used in Google’s image search and speech-recognition software. The ultra-intelligent machine Le once imagined remains distant. But seeing his ideas make software smart enough to assist people in their everyday lives feels pretty good.

    Tom Simonite

    Watch this Innovator at EmTech 2014
    Meet the Innovators Under 35

  • Julie Shah

    Age:
    32

    This MIT engineering professor is turning robots into ideal colleagues for humans.

    “In factories there are usually physical barriers between people and robots. Originally, this was for safetyindustrial robots were unwieldy and unyielding. Although robots are increasingly designed to safely share human workspaces, even in these settings, people do one set of jobs and robots do another. 

    “Imagine if robots could be truly collaborative partners, able to anticipate and adapt to the needs of their human teammates. Such robots could greatly extend productivity. That possibility is really exciting to me.

    “Human interaction isn’t part of the traditional curriculum for training roboticists. Our field is always pushing to make our systems more autonomous, and have richer capabilities and intelligence, but in that push we tend to look past the fact that these systems are, and always will be, working in human contexts. 

    “My lab is now focused on how to create robots that make flexible plans and reconsider their best next action based on changing conditions. It’s a challenging problem, because it’s so hard to model peopleto know exactly what they’ll do and when. You also have a computational challenge, because the robot needs to reason on all these possible futures so quickly (the way humans naturally do so well). And you need to make the robot interact in a way that a person will accept. But experiments show that when people work with the adaptive robots we have designed, they can complete their task faster, with less idle time, and they even feel safer and more comfortable.

    “The interesting thing about this is that there’s evidence to suggest the techniques can translate to better human-machine teamwork in almost any settingfrom manufacturing to operating rooms to military applications. I think the insights will apply very broadly. After all, good teamwork is good teamwork.”

    —as told to Will Knight

    Watch this Innovator at EmTech 2014
    Machines Like Us: Robots and Drones at Work