Can AI Actually Be the Tool That Brings Government Back to Life?

What if the real threat to democracy isn’t artificial intelligence, but a government that never learns to use it?

For 50 years we have been told government is the problem: slow, bloated, hopeless, a relic that could never keep pace with the modern world. Then came DOGE, swinging a chainsaw in the name of efficiency and cutting out much that worked along with much that didn’t, as if the whole apparatus were beyond saving. But what if we have been asking the wrong question all along?

In this WhoWhatWhy podcast, Beth Simone Noveck makes the unexpectedly optimistic case that AI is not just one more thing for government to regulate. It may be the tool that finally reinvents how government works, and more to the point, how it feels to deal with it. 

You can get a tube of toothpaste delivered by dinner, but the government benefit you are owed still takes months, three forms, and a phone tree that hangs up on you. 

Noveck’s question is why institutions built for an age when information was scarce still cannot listen, learn, and respond in a world drowning in it.

Noveck founded the Open Government Initiative in the Obama White House, advised 10 Downing Street and Angela Merkel’s digital council, and became the first state chief AI strategist in the country in New Jersey. 

Today she directs The Governance Lab at Northeastern and runs InnovateUS, which has trained public servants in AI across all 50 states and 80 countries. Her new book is Reboot: AI and the Race to Save Democracy. She has spent her career inside the machine, in the long, frustrating fight to drag public institutions into the present.

Why have so many waves of civic technology promised everything and delivered so little? What did the wreckage of the healthcare.gov rollout actually teach us? 

And can government ever become as responsive as the private sector without quietly trading away accountability, transparency, and the things that make it democratic in the first place?

This goes well past the tired argument over AI regulation. It asks whether technology can rebuild trust in institutions that millions have already written off, and what has to be true when the companies building these tools are private and the stakes are public.

Government has become the waiting room of democracy. Noveck asks the radical question: What if AI could finally open the door?

Apple PodcastsGoogle Podcasts RSS

Full Text Transcript:

Jeff Schechtman: Welcome to the WhoWhatWhy podcast. I’m your host, Jeff Schechtman. For 40 years now in the Great American Project, one thing that the left and right often came to agree on has been learning to hate our own government. Ronald Reagan told us government was not the solution; it was the problem. Grover Norquist said he wanted to shrink it down, small enough to drag into the bathroom and drown it in a bathtub. And this past year, Elon Musk made the metaphor literal, brandishing a chainsaw and turning DOGE loose on the federal workforce in the name of efficiency. Three generations of the same lesson, taught so well that most of us quietly stopped believing public institutions could do anything at all, let alone something new or something good. And the lesson held, in part, because daily life kept proving it right. The last time most of us actually touched our government, it had nothing to do with voting. It was a license that needed renewing, a benefit we were owed, a permit, a record, a form that asked the same question three different ways, and a website that timed out before we could finish answering. We remember the music on hold, the office that closed at four, the letter that said our request could not be processed and never said why. Whatever the speeches promised, this is where most people actually meet their government: in the waiting room. And that waiting room has been a cold and discouraging place for a very long time. My guest, Beth Simone Noveck, has spent her whole career in that waiting room. She has worked in the Obama White House, at 10 Downing Street, inside the German Chancellery, and in the back offices of New Jersey state government. And the thread through all of it is an almost unfashionable conviction that the line can move, that the form can make sense, that an institution can be taught to listen. While the rest of us gave up and learned to get around it, she stayed behind and kept working on the plumbing. Her new book, Reboot, rests on the strange and quiet irony that we have built the most astonishing technology of our age, something that can write and reason and paint and persuade. And we have aimed it at nearly everything except the places where people quietly lose faith. Noveck wants to aim it there. She believes that artificial intelligence, the very thing that so many are bracing against, might turn out to be the thing that finally answers the phone. What unsettles her is that the same instrument cuts both ways. A state that can finally see its citizens clearly is also a state that can finally watch them completely. And nothing inside the technology decides which one it becomes. That part — that part is left to us, and we are deciding it now, mostly without noticing. So the question worth holding on to is which of those two governments we are building while the choosing still belongs to us. It is my pleasure to welcome Beth Simone Noveck here to the WhoWhatWhy podcast. Beth, thanks so much for joining us.

Beth Noveck: My goodness. Thank you so much for having me and for that incredibly powerful and thoughtful intro.

Jeff: Well, it is a delight to have you here. Thank you so much. The idea of bringing technology to government is something that is not fundamentally a new idea. There have been experiments, as you know, far better than most of us, with Web 2.0, with open government, with civic tech — it came with all sorts of names. None of them have really delivered very much. None of them have worked very well. Talk about that first of all.

Beth: Of course, government has been central both to the invention of technology and the adoption of technology, right? It’s government investments that led to the creation of the internet. It was a government grant that led to the development of Apple and, of course, government, although slowly, has been the largest purchaser of technology in the country and used technology for many good things. I would agree that we haven’t done as much as we ought to have. But we have done many things, not just Al Gore inventing the internet, of course, but the adoption and rise of digital government, the ability now to transact with government via the web instead of having to go in person and wait online, surely is an advance. The opening up of government data, something I’m proud to have worked on, the fact that — and again, we’ve seen some retrenchment in this — but for a while, and at least globally, we are making an effort to shine a light onto the workings of government, create more data transparency, publish the data the government collects, again, the taxpayers paid for. However, however… So I do want to say some things are going well, and of course there’s a lot of wonderful people who have invested heart and soul in doing some of those things that have been transformative. But we have not done as well as we’ve needed to, and especially in comparison with the private sector, where I can order a bar of soap or a tube of toothpaste and get it the same day, our expectations have increased for what we need and want government to do, and is therefore a constant and too frequent disappointment when it doesn’t do what it needs to. When we don’t get the services to which we’re entitled, when we don’t have the ability to, and when we sit on the phone or sit on the web and don’t get the answers to our questions. And more importantly, frankly, that we have not reimagined or reinvented or reinvigorated the democratic spirit of our government. We have less voice today, less power today, less input into how government works, I think, than ever before, despite having all of these tools that allow each of us to communicate. So I can make TikTok videos all the live long day, but I really have very, very little influence over, or very little voice or input again, into how government works. So I think, as the question implies, we’re not doing enough fast enough.

Jeff: The other part of the problem is an image problem. And you went through a litany of some things that government has improved in. And I think that everybody listening to us would agree with that. And yet that message never really gets out there, or it never gets out there effectively. Why has government been so bad on the PR side of all of this?

Beth: Well, I think your initial allusion to government being the problem rather than the solution, that message has been peddled for such a long time now. That’s 50 years we’ve been talking about the Reagan-inspired message, the conservative message of government being the problem, need to shrink government. So that has led, and that is not just, I might add, a phenomenon of the right, even if Reagan was the sort of, and Grover Norquist kind of are the originators of those quotes. It is the left as well who said, you know what? The machinery of government, the workings of government aren’t sexy. We don’t want to talk about governance. And this is not just a government issue or a political issue or a PR issue. It’s also a media issue. As the media has gotten more and more concentrated, more and more commercialized, there has been the search for more dramatic headlines, right? We want to engage more listeners. And if it bleeds, it leads. So we love to talk about the sport of politics in the same way that we now want to talk about the World Cup or the basketball finals or other sporting events. We like the who’s up and who’s down. We love to talk about elections, who’s winning and who’s losing, and any conversation about the machinery of government, what happens the day after the election? Frankly, it’s always been viewed as being kind of the third rail of boredom. When I worked on open government for the Obama administration, I was told — quite literally, this is not hyperbole — Obama has to be in another continent before we can make an announcement about open government, because the American people aren’t interested. Well, fast forward, and for all the wrong reasons, DOGE has gotten people really interested in the machinery of government. It’s when we lose it that we then discover how much we actually need it.

Jeff: The other part of the problem is that there’s two fundamental views of government. There’s the broader view of democracy and the founding fathers and the broad base of institutions that people think about on the one hand, and then there’s government in the way that we’re talking about it now, which essentially is a problem to be solved. And in many ways, those two things conflict with each other.

Beth: Well, I think the point is that democracy is a very complicated term. And for most people, if I said the word democracy, if I stopped you on the street and said, what does this mean to you? For most people, it does mean a system of elections. It’s the free, fair, and frequent right to select the leaders who represent us the day after the election. But at the same time, democracy also includes, as you’re pointing out, once we pick those leaders, the services that they deliver, the policies that they make, what happens the next day, and whether we have institutions of governance that are actually capable of solving problems that are effective. And so having effective government, having government that can do the things, whether it’s reduce traffic or deliver health care or school our children, the ability to do those things effectively is what leads to the question of whether we do or don’t trust our institutions. So those things are part and parcel. And of course, then we have the values aspects of democracy, the rule of law, transparency, civil rights and civil liberties. All of these are of a piece. And again, depending on who you talk to and the day of the week, they are all, for different people, these are different. They have a different role to play in our democracy, but they’re all part and parcel of what we mean. I don’t think there’s any conflict, however, between how we think about elections, which itself is and requires and depends on effective and transparent machinery — in other words, the ability to deliver and run those elections, which are frequently being called into question now to a greater extent — and our ability to do other things like deliver a food benefit or reduce traffic or deliver health care, all of those things are part of the administration of governance. And the way in which we then create and run institutions that help us address our societal challenges.

Jeff: One of the other aspects of it, and this goes to the idea of problem solving, is that if we look at so many of these institutions, technology should be a help because basically they’re information processing systems in large measure. 

Beth: Correct, 100 percent. I mean, really, what organizations and institutions do — not just government, businesses, universities — these are essentially bodies that we create to take in information, analyze information, and then output some kind of decision based on that information. And that is traditionally, and that’s how we’ve arranged our government institutions, is around the concept of information scarcity. So we used to have very little access to information. Information traveled very slowly. So we created this concept of bureaucracy, these experts in health and environment, in the affairs of state, in international relations. Who would be the mandarins who have the best knowledge, who have the access to the best and most up-to-date information. They would, in turn, continue to acquire more information. And then, in theory, and that’s how we think about bureaucracy and the way that it works, that would output the best decisions. And then fast forward, we now have these technologies that allow all of us to communicate, all of us to talk, and we are awash in information, and we recognize that we need to have more and better information in order to make better decisions. But we have these systems and institutions and ways of working that are really designed, again, for an era of information scarcity when we’re living in an era of information abundance. And so learning how we operate, how to work in this new way, how can we take in more intelligence and use it more effectively? That’s the struggle we’re having and where I think AI can be the greatest help.

Jeff: I guess the question is, can you change the model without changing the model? That if we have new and different ways to deal with this information and AI being the penultimate example here, how do you then have to look at the institution itself and realize that it’s not just what comes out of it that has to be changed and the way the information is processed, but then we have to look at the institution itself and its viability in this kind of construct.

Beth: Well, I do think that there are ways in which I think about our democratic institutions of the way I think about umbrellas. It’s not a technology that’s changed a whole lot in the last 100 years, or in the case of our institutions, longer than that. So they’re serviceable to an extent, but there’s really room for innovation and changing how we work. So the first generation of adoption of any new technology, and we saw this with the internet, is always about just doing faster or doing slightly more efficiently, potentially what we’ve always done before. And then the second generation, the next step, is to say, okay, yes, we can do the same thing faster, but can we also do things we’ve never been able to do before? So when you invent the car, you can think about it as the horseless carriage. But of course, the adoption of the car does so much more than just… It’s not just a faster carriage. We know the societal implications of individual transport and what that means in terms of freedom, in terms of the notion of childhood, in terms of how we’ve thought about the way families interact. It has huge societal ramifications. So the thing I’m probably most excited about and most hopeful about when it comes to the way we think about our institutions is that we can really transform them more significantly into learning organizations that are far more participatory and that have a greater ability to leverage collective intelligence from many, many more sources. So maybe let me unpack that very briefly, without going on too long. 

So yes, we can use AI to help us do our elections better and keep voting once a year, once every two years, once every four years, as we’ve always done. And that would already be a good thing. What’s much more exciting to me, though, is the prospect that instead of only having a voice once a year at the ballot box, that we could also do what’s happening, for example, in Bogotá, where they are expanding their use of what’s called participatory budgeting, letting citizens have a say in how budgets are allocated, and they’re using new technology to make it possible. In this case, they just ran a participatory budgeting exercise. They heard from 50,000 people in two weeks. That amount of input previously would have been impossible to manage, but with AI, it becomes much easier to sort through that level of information. It’s places like Hamburg, Germany, who have said, we’re going to not only go ask the public how we should do our infrastructure decision making, our environmental and land use decision making. For example, they took an old train station out of use, and they wanted to turn it into a park and a sort of shopping mall, and wanted citizen input into how they do that. So in this case, they built AI tools not just to allow citizens to participate, but to also allow the city to make sense of what people are saying, not in months or years at the cost of tens of thousands of euros, but actually in a matter of hours and days, so that they could actually use that input and, more excitingly, not do that once, but now do that repeatedly so that all of their land use decision making is made with public input, with the input of people who know best what’s happening in the community, what’s happening in the park, who are neighbors of those areas, who have that collective intelligence that then they can leverage. So to me, that becomes a qualitative sea change when we move from seeing something like participation and engagement as a one-time nice-to-have into a regular must-have that becomes the standard way of operating. Those are the things I’m excited about, as you point out, not just doing faster what we do now, but doing something really qualitatively differently than we’ve ever done before.

Jeff: One of the things that also seems like it has to be different is the attitude of citizens, because to a large extent, because of all the negative aspects of this and the bad PR, even with respect to things that have worked, okay, there’s this attitude that voices don’t matter, that individual opinions don’t matter, that the kind of things you’re talking about don’t really work because people don’t have enough input. It seems to me we have to change attitudes about this.

Beth: You are right. And it is, you know, the truth is they don’t matter. And people with money have more influence. People who can afford to hire a lobbyist get heard in the hallways of Capitol Hill to a greater extent, in the hallways of their state capitol. That’s absolutely true. You and I have very little way, even if we participate, for example, in commenting on a regulatory rule, which by law, since 1946, you and I have the right to comment every time a federal agency makes a rule. So whether it’s the width of doorways or the speed of ceiling fans, things that federal agencies tend to regulate, clean air, clean water, you name it, we have a right to participate. But how effective that right actually is, even in an era of new technology, that right has arguably in many ways become less effective, partly because when so many people are commenting, when so many people are talking, nobody gets heard. So when I worked for President Obama, one of the first things we did was to open up, to ask the American public, and we meant it genuinely when we did it, was we asked people, what should the president’s first 100 days be about? What should his initial agenda be? So every new president comes in and says, here’s what my priorities are going to be. We went out and asked people what they thought, and it genuinely was very well intentioned at the time. However, we got 125,000 comments. And this predates generative AI. So as you may imagine, when you get 125,000 comments, number one, and number two, you’re not set up in any way, both technologically, but also from a process perspective. Nobody was used to getting this kind of input. Nobody asked, gee, how might we change the way that we work to account for this? We didn’t really do very much with that feedback, or I would argue we didn’t do enough with it. So that’s the kind of thing where now the tooling can make it easier to do this. It doesn’t mean that everybody will want to do it. No. Does it mean that some people, a few people who may be well intentioned and say, I’d actually like to go out and hear what people have to say? We’ll start doing it. My hope is if we can demonstrate that it’s both effective and efficient, that we will start to see more people replicating these more open ways of doing things. There are, again, many people who want to, and I recently testified on Capitol Hill. It was a congressional subcommittee. Obviously in this point in time, it had a Republican chair and a Democratic minority leader on the committee, so it was people of both parties who were in attendance asking, I think, quite genuinely the question, how could we use new tools to actually hear from the American people? I don’t think they don’t do it only because they don’t want to. For many people, they don’t do it really because they just don’t know how.

Jeff: If you take that to the next step, if broken institutions are going to use AI and use technology to fix themselves, to the extent that those institutions are dysfunctional at the moment or are captured by special interests, you’re basically trusting the patient to diagnose and treat themselves, which is not always a good plan.

Beth: Well, you are right that… I like the analogy, especially people who sit in positions of power, whether they are in an elected position or whether they’re an appointed position, we tend to be good at the things we know. We don’t like to change. We don’t want to… There’s a reason we have the expression, you know, reengineering the plane while flying it. It’s a very, very hard thing to do. And it’s not always the thing you want to do because it’s not what you’re accustomed to. You’re a real expert at your job the way you’ve always done it. So that’s where learning is extraordinarily important and why I’m excited that AI has created an interest in learning and upskilling that didn’t exist before in the public sector. And one of the things you’re pointing to has really been a very long-standing problem, especially at both the federal, the state level. But at the federal level, we have not had legislation about training or learning in government since 1958. Forget AI. Think of all the things that have been invented since 1958. The personal computer, the internet, I don’t know, Impossible Burgers, all kinds of inventions that we have not taken account of in thinking and rethinking how our institutions work. We do not have budgets within government that are devoted to or focused on learning. The learning we were doing, at least at the federal level, was driven largely by an agency called the Office of Personnel Management. It worked on a fee-for-service basis. That meant if you were sitting in another agency, the Department of Transportation or the EPA, and you wanted your staff to take a course in something like human-centered design — a skill that, I might add, every single employee in a company like IBM has had to learn because the CEO viewed it as instrumental to being effective in the business world in this day and age — you would have to pay somewhere between $2,000 and $3,000 for that course. Now, even if you weren’t paying individually, it was your department reimbursing the other department. If there’s no budget for it, you’re not going to learn those skills. You’re not going to learn those new ways of working. So this idea of reengineering the plane while flying it depends on teaching people some engineering skills, giving them a screwdriver and teaching them how to use it, and for that, we have to make learning much more universally widespread and accessible. Back to the AI point, AI — and the sense of the feeling that AI is instrumental to American competitiveness, American economic development, the creation of jobs — has led a lot of leaders, especially at the state level, to say, I want my workers to know something about AI. So Ohio has gone out and trained 27,000 people in AI in their public sector workforce. California is a little anomalous, in a good way, in that it has something called Cal learns. It has had a learning kind of body or agency for a very long time that has been responsible for upskilling the state workforce. There has been investments in learning these new skills and including human-centered design. But California now too, has gone out under Governor Newsom, and issued essentially a call for all public servants to learn something about AI. Again, the tail wagging the dog a bit. We should have been doing more of this sooner, but I’m happy that we’re doing it at all, because that’s the only way that we’re going to reinvent how we do things. It’s not going to just be bringing in new people from outside. We have to also change the DNA from within.

Jeff: We talk about AI as a tool in all of this. We talk about it in its ability to handle and process all of this information. To what extent are we thinking about it as a part of the deliberative process? Is AI in the room part of the deliberation in making these changes?

Beth: So it’s an interesting question, and I’d be interested in more about what you’re thinking here. There have been some people developing tools for AI-enabled deliberation, by which I mean getting AI to talk to AI. We can mean different things by this. There’s then AI-enabled deliberation in the sense of using AI to help do things like summarize or synthesize an argument so that humans having a conversation can do so more efficiently. I’m excited about the latter, not about the former. I am really excited when, for example, the Mexican Supreme Court gets 4,000 citizens sitting down at tables in person, talking about, deliberating about the future of their criminal justice system, where there have been challenges of both corruption and inefficiency, and saying, hey, we know that, we want to do better. Talk to us, tell us how we can do better. Four thousand people deliberating and talking and debating… the help of some tools to make sense of that quickly, summarize, help people themselves understand what are the different positions, who’s on which side of which issue? Where do we agree? Where do we disagree? Some great new tools being developed for helping groups do things like moderate a conversation, achieve consensus. And I think all of us now are familiar with, at the very least, whether you like it or not, your phone is summarizing your voicemails for you. We see the application of AI for helping us to manage conversations. What I’m very worried about is people short-circuiting and shortchanging the human aspect of deliberation. So the idea of saying, I’m just going to have AI talk to AI and come up with a solution. That is, I think, should be self-evident why that’s a terrible idea. I’ll just give you two short reasons. One of which is AI does not have the full range of human experience and human know-how and human creativity that we need to bring to bear when trying to solve real-world problems. AI is trained on written literature and even written literature, things like journal articles and things on the internet, do not give you what you need when you’re trying to decide, what should we do? What’s the future of this hospital? Or how do we think about diabetes care in our neighborhood? Or what do we want to do with this playground? Human experience is vital, vital, vital to this process. And second of all, decisions that we make have to be not just effective, they have to be legitimate. For this to be a democracy, we have to have some engagement or involvement or sense of voice. And if we’re just going to let AI deliberate with AI, we’re going to be worse off than we are today and just compound the problems you’ve talked about earlier, the lack of voice, the powerlessness and the lack of accountability, especially which we probably don’t have time to get into, if these tools are not controlled by us, not run by us, not owned by us, but entirely private.

Jeff: Which raises the question of really where I was going with this is who does all this? Where does the procurement come from? Who’s responsible for making these sea changes? Because that has a lot to do with outcomes as well.

Beth: So I’m glad you raised the word procurement. Again it’s another one of those words that we dare not speak its name because it’s generally considered so dull, right? Nobody wants to talk about the rules around which institutions buy things. Sounds incredibly wonky and bureaucratic, but that is really where the rubber meets the road. Again, with government being the largest purchaser of technology in the country and in the world, it has the ability to shape markets. It has the ability to shape how these tools run. So if we complain and we worry about these tools being unaccountable, these tools using our data, these tools robbing us of our privacy, us not having a voice in how they’re created, how they work, we don’t have insight under the hood around what’s happening, all of that can be in part addressed not just through legislation — we have lots of discussion about regulating or legislating around AI — we can address it much through procurement, through how we buy these tools. And if we had only pushed back sooner before purchasing some of these other things, we would have better tools. And government has long done this, it has been the maker of markets. Government during COVID said, we will buy millions of doses of a vaccine if you go out and make it. So the companies went out and made the COVID vaccine and accelerated their research because government made what’s called an advance market commitment, said, we’ll buy it if you come. And we need to do the same thing around our AI, is to say, we want tools that are transparent, that are open, that are affordable, that are designed for public purpose, that are optimized for use in promoting democratic engagement, and then we will buy it. And that’s a power that government has. And it doesn’t take a lot to do that.

Jeff: It seems that on one hand, within that framework, we’re substituting consultants and McKinsey people for lobbyists who we had before.

Beth: Well, the government is swarming with them, obviously, because again, it’s the trough around which people feed because there is so much money, and that is one of the challenges. And because it’s such a significant buyer and user of things like technology and other services, but it’s also where the power comes in. And that’s why I’m so excited about the opportunity for learning, because it’s only by learning both what these tools do well, what they don’t do well, what they could do, that we’re going to get government officials who know how to push back and are informed enough to do so.

Jeff: Where can all of this go wrong?

Beth: Well, I think we’ve already alluded to one thing, which is overdependence on these technologies, number one. And we see this as not just a question of government and democracy. We’re wrestling with this question as we think about how to roll these out in our schools, right? Schools debating now whether to ban the use of phones or technology. Do we ban or keep the use of AI? One of the big debates is, are we going to become too reliant on these tools and on the wrong tools? So one risk is just, again, that we do some hard learning and thinking about how do we incorporate these into how we work. 

The second is, of course, overreliance on tools that are accountable to nobody. And we’ve seen what’s happened with AI that can hallucinate. AI makes stuff up, but also AI can be used to push a particular point of view, especially if it isn’t publicly managed, governed, and regulated. You can have tools that start spitting out white supremacy, that start spitting out lies, that start spitting out propaganda. So as we bring these things kind of into our lives, it can go really, really wrong. I’m afraid if we do not have more publicly governed and managed alternatives to the system we have now and again from an individual perspective — so that’s the sort of institutional perspective, the individual one is that I think all of us are grappling with, as we did with the web and the internet 30 years ago, with the question of what does this mean for how we work? What does it mean for how we learn? What does it mean for how we do our societal relationships? And I think we’ve made some missteps in the past and we’re paying the price now in terms of loneliness, in terms of the echo chamber and people shouting at each other, in terms of the political divisions we see in this country that tech has exacerbated. So I think we’re doing some hard thinking about this.

Jeff: And this brings us full circle with respect to the concern about government getting involved in this, because people look back to what happened when the ACA went online.

Beth: So you’re 100 percent right… but that was a good example, though, of the expression, “Never waste a good crisis,” or “Never waste a good scandal.” What happened there is that, of course, things didn’t work. And that created a lot of attention to this concept of a website that’s broken, number one. Number two, the recognition of how important implementation is, not just policy. So we can have a policy that says we’re going to give health care to people, we’re going to create infrastructure for people, we’re going to do X, Y, Z. The policy is really 10 percent of the battle, 90 percent is the implementation of that policy. So I can have a mandate around health care, but if I don’t have a website where you can sign up… I can have a mandate around infrastructure, but if I can’t build a bridge… it’s literally where the rubber meets the road. And so that is where we’re seeing, especially after DOGE, a lot more attention to these questions of what some people call capacity or implementation, our ability to actually get stuff done. So we see the problems of, for example, climate change and rising temperatures. We see the fact that 40 percent of kids read at grade level. We know how to fix these problems. But where we get stuck is sometimes in this question of, from the good intentions and the ideas and the policy, into the implementation. It’s people who do that implementation. And that’s why making sure they have the tools and they have training that can help them get more effective at doing these things is so crucial.

Jeff: Beth Noveck, her book is Reboot. Beth, I thank you so much for spending time with us here on the WhoWhatWhy podcast.

Beth: I thank you for these incredibly thoughtful and intelligent questions, and I greatly appreciate the conversation.

Jeff: Thank you. And thank you for listening and joining us here on the WhoWhatWhy podcast. I hope you join us next week for another WhoWhatWhy podcast. I’m Jeff Schechtman. If you like this podcast, please feel free to share and help others find it by rating and reviewing it on iTunes. You can also support this podcast and all the work we do by going to WhoWhatWhy.org/donate.