A Better Tomorrow: Renewing R&D’s Promise to America
Adapted from the AAAS Carey Lecture delivered June 21, 2018, in Washington DC
The purpose of science and technology — of research and development — is to enable a better future for society. Today the U.S. R&D ecosystem looks like it’s booming if we count papers, patents, and PhDs. But if we ask what our country and our world need to thrive in the decades ahead, it’s clear that the half a trillion dollars we spend annually on R&D isn’t focused on preparing us for the future that’s coming. The gaps in our current investment portfolio are signals that it’s time to take a full-system look at U.S. R&D and do the creative experiments that can lead to new and more effective ways of working.
My views are shaped by the many different vantage points I’ve had on R&D and its role in our society. I’ve had the privilege of leading two wonderful and very different public organizations: the National Institute of Standards and Technology (NIST) in the mid 1990s and the Defense Advanced Research Projects Agency (DARPA) during 2012–17. In between, I worked for 15 years in Silicon Valley, first in a couple of companies and then for a decade as an early-stage venture capitalist. My first job after getting a PhD at Caltech was as a Congressional fellow at the Office of Technology Assessment, and I just completed a year as a Berggruen fellow at Stanford’s Center for Advanced Study in the Behavioral Sciences (CASBS). In several of my roles, I’ve had the chance to work with many dozens of other R&D organizations — universities, government labs, nonprofits, and companies of every size and type — and to build teams to advance research, bring new products to market, and demonstrate powerful new technological capabilities to serve a diverse set of our society’s needs.
As I look at our U.S. R&D community as a whole, over a period of many decades, it’s clear that we have been very successful in many ways. But the challenges for the future are different than those we’ve faced before. And the case I want to make now is that we will need new kinds of advances and new ways of working if we are to contribute for the next handful of decades as we have for the decades just passed. That’s what it will take if we are to renew and fulfill R&D’s promise of a better future for our country and our world.
It takes an entire ecosystem
Let’s start with how things are working now — because the work of scientists and engineers and technologists has achieved a tremendous impact for our society over the last handful of decades. Let me pull just one thread from that tapestry for an example.
Every time I look at the blue dot on my map that tells me where I am, I remember a couple of young researchers at Johns Hopkins’ Applied Physics Laboratory. During the excitement and panic of Sputnik, these two lugged their radio gear up to a hillside where they listened to the Doppler shift from that first satellite that humans had ever flung into orbit. GPS is a direct descendant of their observations 60 years ago. And of course today, it’s a ubiquitous utility that has transformed how we grow our crops, how we orchestrate trucking and logistics, how we get across town in traffic, and even how we keep track of our kids when they’re young.
To get here, it took billions of dollars and decades of investment by the Defense Department. It took two Presidential orders that opened GPS signals to public use. And it took a long series of companies and entrepreneurs and investors to put GPS receivers into our phones and computers and trucks and tractors and cars — and then to create location-based services. That’s how those little blue dots appeared on our maps.
GPS is just one example of the powerful technologies that have changed our lives and our society and our world in the last few decades. Each technology has its own story, whether it’s advanced materials, new energy technologies, medical therapies, or any of the waves of the information revolution, from the integrated circuit, to the internet, to wireless and mobile technologies, to data analytics and artificial intelligence. These different technologies germinate, grow, blossom, spread in every conceivable way. But they share this characteristic: every one of these powerful technologies was the result of our wonderfully complex, convoluted, robust research and development ecosystem and the people who inhabit it. Each one required basic research, each required product development, each required the interplay between those, and each required the interplay of R&D with markets and societal needs.
Looking to the future: the purposes of R&D
The question now is this: What does that ecosystem need to be for the future we want now? The place to start is the purposes behind the investment in R&D.
The United States spends over half a trillion dollars on research and development each year — public and private — from the most fundamental explorations in a university lab to incremental improvements for a product that’s already in the market. And in many ways, things are roaring along, with new advances coming into our lives at an astonishing pace.
But we also have some work to do. Our nation faces some significant challenges. Here are five areas — five pillars of our future well being — where we are not prepared to meet the future.
Economic prosperity. In terms of overall economic growth, real U.S. GDP growth hasn’t exceeded 3% for a dozen years. And in terms of individual economic opportunity, today we live in a country in which 44% of adults say they couldn’t cover a $400 emergency. We live in a country where in 99% of our neighborhoods — Census tracts — black boys grow up to make less than white boys who grew up in families with comparable income. How will we expand our economy in the decades ahead AND live up to the vitally important American ideal of being the land of opportunity for every person?
Our democracy. At a time when democratic norms are eroding in many places around the world, social media and online networks are introducing new vulnerabilities to our democratic discourse. The reach is vast: last year, 2/3 of American adults were on Facebook, and 2/3 of them got news — including “news” — that way. That’s 45% of American adults. How will we address those vulnerabilities and recapture the ideal of online connectivity bringing us together rather that driving us apart?
National security. Unlike in the Cold War, today, with a few important exceptions, we in the United States build our military capabilities from the same technology building blocks as the rest of the world — with enormous talent but at a pace that will not suffice to deter and defeat many aggressors. How will we adapt our military capabilities faster than any current or potential adversary?
Health. We spend substantially more per capita than any other nation but have lower life expectancy than 26 other nations — and our healthcare costs continue to rise faster than our GDP while life expectancy has dropped for the last two years. At $3.3 trillion and 18% of our economy, growing healthcare costs bite every company and every individual — and rising healthcare costs are also eating the federal budget, putting massive pressure on all discretionary spending (including, by the way, federal R&D spending). How will we achieve better health outcomes while significantly cutting cost?
Basic human needs. For those of us who are doing all right in developed regions of the world, we are fortunate enough to have forgotten what it’s like not to have ready access to clean water, and plentiful food, and cheap energy. But as we start into an extended period of climate instability, no one can take these basics for granted. How will people here and around the world meet their basic human needs in a time of climate instability?
Now, science and technology won’t solve any of these problems by itself — it isn’t sufficient. But it is necessary. None of these daunting challenges will be met without research and development. And while our current R&D efforts are doing a lot, they are not fully stepping up to these important challenges.
What new ways of working and what new advances will help us address these challenges? Where are the fruitful areas to experiment, to try new approaches and assess their impact, and then to stop what doesn’t work and scale what does so we can improve our odds for the decades ahead?
I want to give you two very different examples — two different areas where the time is ripe for this kind of experimentation. One is about improving the odds for successful research to move to commercial and societal impact, and the other is about tapping the growing potential of the social sciences. These ideas are by no means exhaustive. The challenges ahead are vast and diverse, and they will demand a wide array of new approaches. And the work ahead will require and benefit from the support and participation of multiple players in our R&D ecosystem — companies, philanthropies, government, universities, and individuals.
Achieving the benefits of research insights that are stranded in the lab
The first area is about making it more fluid, more possible for promising new research insights to advance to the stage of commercializability.
In between two vibrant, highly functioning parts of the American system, something isn’t fully working. One of our tremendous strengths is that we generate extremely high-quality basic research in essentially every field. And a second tremendous strength of the American model is a market economy that is astonishing good at applying talent and capital to commercial opportunities.
Sometimes the gap between the realm of research and the realm of business is manageable. For example, for a research advance that can be instantiated in code, it might take a year or two and a million or a few million dollars to build a first product and reach a first customer, so you can start seeing if your business idea really is as good as you think it is. Today, we have companies and venture capital that step up to take that amount of risk as a first step to reaching for a big payoff. And that’s what bridges the gap between many software research insights and commercializable technology.
But when a new research insight is based in material science, or chemistry, or biology — the natural sciences that deal with the physical world — that gap is often insurmountable. The work that’s needed doesn’t fit in the world of peer review, publications, and tenure, and the cost and the time it needs put it beyond the risk profile of private capital. I saw this repeatedly in my time as a venture capitalist.
So what do we do about this pernicious gap — especially for the natural sciences and engineering?
One part of the answer lies in a growing number of efforts that are underway across the country to experiment with directly supporting researchers to take those critical first steps beyond basic research, to develop to the stage of a commercializable technology. Cyclotron Road in California and The Engine at MIT are two of these. One instance of the kind of work is at a new company at Cyclotron Road. It’s in the field of metal organic frameworks — these extremely porous materials that we know from many labs can pull water out of the air, pull CO2 out of the air, do very efficient chemical separations, but that so far haven’t made a dent in any of those problems. This Cyclotron Road team is going from the tiny amount of powder you need to write a paper in a highly regarded journal to the kilograms you need to understand the process cost and robustness and to demonstrate a product idea for a potential customer.
And by the way, the excellent article that reviewed the state of process technology for metal organic frameworks last fall came from South Africa and China. That’s a great reminder that we no longer have the luxury of pursuing these capabilities whenever we can get around to it. If we want our country to participate fully in the benefits of global research, if we want the variety of companies and industries and jobs, if we want the deeper understanding that comes from making and scaling — the further insights that lead to new markets and growth — it’s not enough just to hope to be able to buy the products from elsewhere. We need to get moving to make it much more feasible to commercialize these kinds of technologies in the United States.
And I want to be clear here. I’m not talking about isolating America from the rest of the world, or thinking we have to win in every situation and other regions have to lose. I’m simply talking about the United States being a full and active participant not just in global basic research but also in its application, its scale up, its use. Because as we do that, and as we reap the broader benefits for our country, we also contribute to the global good. That’s how a strong country plays its role in the global community.
Now today, all of these examples of new approaches that could improve commercialization are at a pretty early stage and a small scale — and as we learn what works, they will require focus and resources to scale up. It will be well worth the journey if we can realize new contributions to economic growth and spur new types of jobs. It will be well worth it if we achieve new solutions for national security and new ways to grapple with carbon and energy and water. That’s what we can unlock.
Social sciences for powerful technological capabilities
That’s one broad area with enormous potential. Here’s another. This one is about developing new methods in the social and behavioral sciences and applying them to problems that we’ve long thought of as intractable.
Something exciting is happening in these fields today: we’re seeing some very interesting signs of revolutionary new capabilities across a wide variety of applications.
Gary King and his colleagues at Harvard downloaded billions of posts across 1300 different social media sites in China. By analyzing which ones disappeared, they were able to determine that the Chinese government doesn’t censor criticism, as many had thought. Instead, they found that it very quickly and effectively censors any discussion of collective action — a very important result to understand the nature of the control that’s happening in that society. Now King and Nate Persily and others are teaming up to analyze the impact of Facebook on our elections.
Zoran Popovich from the University of Washington and a nonprofit called Enlearn is advancing the representation of learning content to link it to a student’s thought process — which enables them to develop adaptive learning pathways, on the fly, for each unique child.
Raj Chetty at Stanford is mapping and mining a wealth of data sources to build models that explain the multiple factors in income mobility — how do the factors in a child’s life, like family income, race, family structure, and community, correlate with that child’s movement from one income percentile to another as she or he grows.
Francesca Salipur, Kendell Canon, and Myra Altman — a team of doctors at Stanford who are committed to improving patient outcomes — have identified successful, practical improvements in end-of-life care that if scaled can reduce costs by many billions of dollars.
Opower is a company that has worked with utilities to give households information about how their electricity consumption compares to their neighbors — leading to a surprisingly significant reduction in consumption at a quite modest cost to the utilities.
If you step back, you can see in these very different examples some of the promising signs that social science is entering a new phase. It’s characterized by voluminous data — noisy, messy but phenomenally valuable in its sheer quantity and freshness; the ability to build models of hugely complex systems; and the ability to scale solutions, to personalize them, and to iterate and adapt them to reach more people more effectively.
A couple of big advances have to happen if we are to realize the full promise of these new horizons in social sciences. The first is about security. The most interesting data in the world is about human beings — and that can also be the most dangerous data. We already know what’s at stake: our privacy and individual autonomy, some of the liberties we hold most dear. The good news here is that new mathematics and cryptographic methods are making it possible for the owners of data to be sure that their data is used for the purposes they agree to — and no other. These security technologies will need to be applied with thought and care to unleash the value of personal data while protecting our deepest values.
The second advance that’s needed is to open the capacity for innovation and change in our social institutions. Companies adapt and change because it’s do or die. The same is true for national security — literally — so we have built in R&D and experimentation in the Defense Department. But the capacity to innovate and adapt wasn’t built into the design of our schools, our hospitals, and our public organizations. So as before, we need to pay attention to the entire ecosystem, not just the research component, if we are to succeed in reaping the many benefits of advances in the social sciences.
The ethical implications of powerful technologies
Such exciting times…. Here we have the social and behavioral sciences converging with the information revolution and math and crypto, exhibiting all the signs of the birth of a new set of powerful technologies. And that makes this a good point to take a minute and ask: “What could possibly go wrong?” It’s an essential question whenever we’re talking about powerful technologies — of any kind.
We can look back again at the example of GPS. Along with all of its wonderful commercial applications have come new questions about privacy and who has the right to know and to use your GPS data. And in the Defense Department, where this particular powerful technology started, GPS was a key enabler of the biggest strategic shift in military technology in the last half century: the shift from mass to precision. That’s dramatically reduced collateral damage, and our use of stealth and precision weapons in the first Gulf War was so overwhelming that it arguably has shaped the behavior of Russia, China, Iran, and North Korea for over 25 years. And at the same time, precision has arguably also reduced the barrier for entry into armed conflict for a series of Presidents and senior military leaders. It’s the old story, the bright and the dark implications of technological advances.
Because technology’s power is a raw kind of power. History shows that it has been used for both good and ill. And history also shows that when individuals and societies make smart, ethical choices about how to use them, powerful technologies — over time and in aggregate — have lifted up our communities, our nation, and our world. That’s why we engineers and scientists do this work. This is our extraordinary privilege.
It also means that we have some extraordinary responsibilities. We need to examine our choices in research and development. We must be clear with ourselves and with users about what emerging technologies can and can’t do. And we need to play our part in helping our society make wise choices about the fruits of our labor.
This has always been the case, but it’s more important than ever now. Today’s generation of advanced technologies don’t sit far away in factories or specialized facilities. Information technology, biological technologies, social technologies — these technologies are intimate with us, so thinking through their ethical and societal implications is more vitally important than ever.
It’s our time now
Seven decades ago, just as the Second World War was ending in 1945, Vannevar Bush wrote “Science: the Endless Frontier”, the blueprint — still — for federal support of research. He too started with national purposes and national needs. He wrote of improving health and longevity. He wrote about military preparedness even in peacetime. And most poignantly, after the Depression and years of combat, with millions of GIs about to come home, he wrote, “One of our hopes is that after the war there will be full employment, and that the production of goods and services will serve to raise our standard of living.” Then Bush laid out new ways of working to help achieve these ambitions. Most notably, he introduced the idea that university basic research is a vital part of achieving the nation’s goals.
Imagine if Bush could walk in our world today — if he could see what’s come from his work. I like to think he would simply beam with delight. The societal impact and the outcomes from the concepts in “Endless Frontier” have been spectacular.
But Vannevar Bush wrote that document in a very different time. It was a time when U.S. GDP was over half of the world’s total production, as the rest of the world was buried under the rubble of war. It was a time when manufacturing and agriculture were the backbone of our economy. It was a time before the first artificial satellite, before the transistor, before we knew that DNA is a double helix. Our opportunities now are different, our challenges are different, and our global context is different.
Now it’s our task to push forward, to explore and invent the new approaches that can address the future we face. Now it’s our time to renew R&D’s promise of a better tomorrow.