JOSH: Hello, and welcome to Clearer Thinking with Spencer Greenberg, the podcast about ideas that matter. I'm Josh Castle, the producer of the podcast, and I'm so glad you've joined us today. In this episode, Spencer speaks with Bradley Tusk about political incentives in gerrymandering, election technology and fraud, and free speech and innovation.

SPENCER: Bradley, welcome.

BRADLEY: Hey, thanks for having me.

SPENCER: So what's the reality of how politicians actually get elected?

BRADLEY: The reality is very, very few people end up deciding who gets elected. And even worse, they also decide what their policies then are once they're in office. Because of gerrymandering, which is the process where the parties divvy up all the different legislative districts to say, "This is a safe Democratic seat, this is a safe Republican seat." As a result, the only election that typically really matters is the primary, and primary turnout in this country is typically about 10 to 15%. So we're talking very, very few people actually participating in our elections. One, that means very few people decide who wins the primary. Two, once you are the politician, all you know is that, "Hey, I want to win my next election. I want to stay in office." And so you've got to keep that small group of voters happy in order to win your next election. And the problem is that ten to 15% tends to not be representative of the people as a whole. It tends to be the extremes: the far-right, the far-left, or special interest that could move money.

SPENCER: Let's unpack that a bit. First of all, for our international audience, we're talking about US elections here. It might be different in your own country. When you're talking about gerrymandering, is the idea essentially that people are changing district lines to make it so that, in a particular district, it always goes red or it always goes blue?

BRADLEY: Exactly. What you literally have — and I've been part of this process in my time in politics — is the party leaders of both parties will sit down and they will literally divvy up the map and say, "This one is mine, this one is yours."

SPENCER: Collaboratively, like they're working together to do that?

BRADLEY: Sometimes they fight each other and sue each other, but a lot of times, they're all in on it together because, ultimately, they just want to make sure that they stay in power. And staying in power is more important to them than any particular ideology, any particular issue, any particular policy. What we have in this country is a small group of people with a lot of power who will do anything to hold on to that power, and gerrymandering is a good example.

SPENCER: My understanding is that gerrymandering is usually legal. Are there some limits to it?

BRADLEY: Yeah, there are some. It does have to pass muster with the Justice Department to be in accordance with the Voting Rights Act. Sometimes, if they determine that the maps that are drawn by legislative bodies discriminate against particular racial groups or other protected classes, they may reject it and send it back to the drawing board for that. That does happen, but that's at the extremes. By and large, it is sadly legal, and unfortunately, the Supreme Court has made it clear that they're not going to change that, which means we are living in a world where gerrymandering isn't going away, which means the primaries — we believe the relevant election — the question then becomes, how do we make the primary the representative of the people as a whole so our politicians are encouraged to actually work together?

SPENCER: Yeah, and if you search online, you can find these absolutely ridiculous voting maps where it's clear someone just drew with a pen, "Oh, I want this group of people to be voting in this district." They just make no sense.

BRADLEY: They do block-by-block, literally, like, "Oh yeah, my friends are on this block, but not on that." And some of them look like crazy jigsaw puzzles because it's not about the representation of the public or democracy. It's purely about people in power keeping power.

SPENCER: The second point you made, I think, is really important, which is that, once things are gerrymandered, what that means is you're not really in competition so much with the other party. You're just in competition with your own party. And so what matters therefore is the primary. Let's talk about what actually determines who wins the primary.

BRADLEY: What determines it is very, very few people. For example, you and I both live here in New York City. The city council primaries were last year. Do you know what the turnout was? Just take a guess.

SPENCER: I imagine it's appallingly low. I didn't vote so I'm just as guilty as anyone else.

BRADLEY: 6.5%. So effectively, in a city of 8.5 million people — in what I would argue (and hope you would agree) is the greatest city in the world — you could win a city council seat with six, seven thousand votes. That is, in no way, representative of the city as a whole and I'll give you a tangible example. This year, the city council passed a bill that said that, every time a police officer interacts with a member of the public, they have to write a report. And that could be whether it's a law enforcement issue or giving a tourist directions to the Metropolitan Museum of Art. As a result, what that's really intended to do is tie up the cops with endless paperwork. Since defunding the police has not become politically salient, it's sort of a backdoor way to take cops off the beat and instead just keep them busy filling out reports and things like that. The mayor vetoed the bill, but the city council opposed this veto. The reason why the city council passed the bill and overturned the mayor's veto is because, at 6.5%, the people who vote in their primaries are the furthest of the furthest Left here in New York City, and they genuinely believe that there should not be law enforcement, there should not be incarceration, there should not be private property, there should not be prosecution. And we have the same problem, by the way, in places like Texas, where it's the far-right and it's the exact same thing. Neither of those things is good. And so we live in a world right now where having incredibly low-turnout primaries determine everything. Either it gets us the complete polarization dysfunction of Washington, DC, where nothing can get done, or totally one-sided government, whether it's the state of Texas on the Right or the city of San Francisco or here in New York City on the Left. And none of that is good.

SPENCER: It's so interesting how you don't actually need that many people to vote as long as the people voting are representative of the population. Ten thousand people could do a good job of representing even a large population if they're selected at random. But I think the problem you're pointing to is that it's an extremely non-random sample. It's a very unusual group of people voting in the primaries, which means that, rather than having an incentive to represent everyone, politicians have an incentive to represent a very unusual group with very specific opinions.

BRADLEY: Absolutely. Let me give an example here in our home city. Do you remember when Amazon wanted to build their second headquarters in Queens a couple years ago?

SPENCER: I do, yeah. And it got blocked, right?

BRADLEY: Right, it got blocked. What's interesting is Amazon wanted to put it in Long Island City which — for those of you who don't know New York that well — is a part of Queens, right across the river from Manhattan. And when you looked at the polling, the vast majority of the city supported the deal. The vast majority of the district where the project was going to be supported the deal. But turnout in that state senate primary is typically around 10%. AOC had the neighboring or congressional district kind of overlap in that area, and when she came out against the deal and said, "No, I don't like this," the state senator — a guy named Mike Geniaris — said, "Oh, no. If I support the deal, in my next primary, AOC is going to launch someone against me from the Left and I'm going to lose and my job is going to be over." So even though Gianaris knew that, not only did most of the city want the deal to happen, but most of his own constituents wanted the deal to happen, he also knew the people who actually show up and vote in this primary would punish him for it. And he was faced with a choice: 40,000 jobs, which had been all the new jobs created by the project, or one job, his own. And he picked his one single job over 40,000 new jobs for New Yorkers. And it's not that Mike Gianaris is stupid; he's not. It's not that he's evil; he's not. He's just a politician. And politicians, in my experience — and I have worked in city government, state government, federal government, as Mike Bloomberg's campaign manager for mayor of New York, I was the deputy governor of Illinois, I was communications director of the US Senate — I've seen this from every angle. Politicians make every decision solely based on the next election and nothing else. So if you want them to do what's right for the most people, you have to align our policy centers with their politicals.

SPENCER: Do we have a sense of what percentage of positions people are running for in office have this problem where it's not really going to be contested by the other party, and so it really ends up just being about the primary?

BRADLEY: The vast majority. For example, if you take Congress, there are 435 seats in the House, and about 25 of them at this point are true swing districts. That is roughly 5% and the other 95% are all effectively gerrymandered to determine the primary. And you know what? It stinks for everyone. My sister's husband — my brother-in-law's name is Josh Gottheimer — and Josh is a congressman from New Jersey, and Josh is incredibly lucky because he happens to represent one of those 25 swing districts. As a result, his election isn't determined in the primary. It's determined in the general. And for Josh, what does that mean? It means that, not only is he not punished for working with the other side; he's rewarded for it. So Josh created a group called the Problem Solvers Caucus in Congress that brings together moderate Democrats and Republicans to try to work together to get things done. Josh has an incredible luxury and privilege that the vast majority of his colleagues don't have, simply because he has the freedom to work with everyone. If everyone had that kind of freedom because they weren't stuck with tiny amounts of primary voters deciding their fate, then they could work together, and we could find compromise, and we can get things done, whether it's climate or affordable housing or education or inflation or guns or whatever issue you want to pick. The reason why we can't get things done is because we're totally polarized, and we're polarized because we have incredibly low-turnout primaries.

SPENCER: To what extent is it really gerrymandering that's causing this versus just natural grouping of people? Because it makes sense that people who have a similar political persuasion will tend to live in similar areas.

BRADLEY: Yeah, it's a little bit of both. But I think, in reality, you could say, "Hey, people who choose to live in places like New York City are more likely to be Democrats. People who choose to live in rural Georgia are more likely to be Republicans." That's fine, but the problem isn't even really Republican versus Democrat. It's the extremes within it. So the far-left of the Democratic Party is very different from a moderate Democrat. Bernie Sanders is very different from Joe Biden. Donald Trump is very different from Mitt Romney. And so the problem isn't just that each party has radical views; it's that the radical fringe and wing of each party are the ones actually voting in primaries and therefore calling all the shots. So even if we just went up from (say) 12% turnout to 36%, by definition, that will change everything. So at Mike Geniaris's state senate primary, if the turnout was 30% instead of ten, just because he wants to stay in office, the politically smart thing to do would have been to support the deal, which means those 40,000 new jobs would have come to New York. Or let's pick on Republicans for a second. You're a Republican congressman from (say) Florida. Turnout at your primary is 12%, it's just gerrymandered so primary matters. Half of those 12% are NRA members. You may know intellectually that it's crazy that someone could just walk into a store and walk out with an assault weapon, but you also know that, if you were to say that, you would automatically alienate all those NRA members, and you would lose your next primary. Now let's expand primary turnout from 12% to (say) 36%, just based on the polling of most Republican voters, they would say, "Okay, we should have some sensible rules around guns. We don't want school shootings. We don't want shootings in churches. We don't want shootings in Walmart. And therefore we are going to have some reasonable requirements." If you were to have just an even slightly larger primary turnout, you would give those politicians the political incentive and cover to move towards the middle, where things could start to get done.

SPENCER: It's obviously not always the case that Moderates have the right answer to everything, but I do think there's an interesting heuristic where Moderates are less likely to destroy things. If you're on one fringe or the other, whether it's Right or Left, you're more likely to push for something really intense that reshapes society in some way, like for example, trying to ban all abortion, or trying to absolutely remove all police, or something like this.

BRADLEY: Right. It's funny. Two things: one is, abortion is a great example. It is the most divisive issue we have, and yet it's really not that divisive in the sense that two-thirds of Americans support the right to an abortion. Now you can debate whether it should be eight weeks or 16 weeks or 24 weeks, that's fine. But fundamentally, it's not like we're 50-50 on the issue. It's that the people who actually vote in primaries tend to be extreme on it so it seems like this incredibly binary thing when it's not. And one of the things that is almost ironic to me as I'm trying to build this movement for mobile voting is, it's a radical movement to move things to the middle, and we're going to have to overcome the powers that be in both parties and the entire establishment not to enact radical policies at the far-left and the far-right, but just to get politicians to start representing the mainstream again.

SPENCER: Just to get this out of the way, sometimes people claim it's just the other side that's causing a problem. It's the other side gerrymandering. Do we see this happening to a very large extent, on both the Left and the Right?

BRADLEY: They're equally terrible, yeah. I've worked in politics in New York, in Washington, in Illinois, in Pennsylvania. I've worked on campaigns — whether it's legislative or electoral policy — in all 50 states. Everyone's the same. Fundamentally, what I have learned with a handful of exceptions, the vast, vast majority of politicians are desperately self-loathing, insecure people that can't do without the validation of holding a public office. There's literally a hole in their psyche, and the only thing that fills it is by being somebody, and for them, that means being in office. They're never, ever going to risk that psychological affirmation that they need just to try to do something good for the world. And so it doesn't matter if you're Republican or a Democrat; the sad thing is, they're basically all the same.

SPENCER: In your book, The Fixer — which I really enjoyed, by the way — you talk about how you worked with a number of politicians, and you worked with companies that were trying to work within regulations and so on. And you laid out this little framework for thinking about different types of politicians, about what the different motivations and drivers are. Now when I hear you talk, you actually seem even more cynical than you did back then.

BRADLEY: Well, I'm pretty sure in The Fixer — I think I said 5% of politicians are exceptions; that might have been generous [Spencer laughs] — but whether it's one or five, we're still talking very, very few.

SPENCER: Yeah. If I recall just from memory, I think that you identified five types of politicians based on your experience working in politics. One, narcissists, which you said were by far the most common type; I assume you still agree with that?

BRADLEY: Yup, absolutely.

SPENCER: Narcissists are attention-seeking and grandiose and they have a certain insecurity, but it's all about their ego. And then you talked about corrupt politicians who are just in it to line their pockets. Once they're in office, they're just trying to steal money or get bribes and so on. Thankfully, in the US, I don't think we have too many of those; although, they're rampant in some countries. Would you agree with that?

BRADLEY: Yeah, I think that that's true. We certainly still have plenty of corruption. Not long ago, Bob Menendez, US senator from New Jersey, was just convicted. He literally was getting gold bars from Egypt to promote their policies in the US Senate. Just yesterday, the New York city schools chancellor, the police commissioner and the deputy mayors for public safety and operations, their homes were raided and their phones were seized in a corruption investigation. So we do have it. But no, this is not widespread.

SPENCER: Right. And then the third type you talked about are ones that really don't want to do any work. I can't remember what you called them.

BRADLEY: The backbenchers.

SPENCER: Yeah. Could you just briefly talk about your experience, because weren't you essentially the shadow mayor? [laughs]

BRADLEY: Oh, I was accused of that by The New York Times. In 2021, I supported Andrew Yang and his campaign to become mayor of New York City. And because Andrew didn't really have any history in local politics or government, our opponent said that secretly, the campaign was designed to make me the shadow mayor of New York City, and that I would secretly be running the city from, I guess, my conference room or wherever they had in mind. One, that wasn't the case. But two, it didn't matter, because we lost anyway.

SPENCER: You know, what I was thinking of is actually you being a shadow governor, not shadow mayor.

BRADLEY: Oh, yeah, I was also accused of that. That was actually more true, yeah. I was Deputy Governor of Illinois from 2003 to 2006 and I ran the state's budget, operations, legislation, policy, and communications, and I worked for a guy named Rod Blagojevich who, if his name rings a bell, it's because he's the guy that tried to auction off Obama's senate seat and ended up with a 14-year jail sentence as a result. Luckily, I was long gone before that ever happened. But Ron had both a very crazy and, in some ways, very logical view of his job, which was his job in his mind was to run for office, not to hold office. And he literally would say, "I did my job, and I'll see you guys again in four years, and I'll do it again." And he did not believe that actually being governor was his responsibility, and he would not come to the office for months at a clip. And I learned this. I wasn't from Illinois; I was living in New York. I was working in city hall with Mike Bloomberg, and I got this call saying, "Hey, you want to be the Deputy Governor of Illinois?" I said, "Why are you calling me?" "Well, it's the person that runs the state," and I end up getting this job. And at the end of that legislative session... Typically, the legislature in Illinois — and it's true of most states — they passed around (call it) 500 bills a session, and 400 of them are nuts, like, "The official amphibian of Illinois is the frog," or whatever. But there's a hundred that have real policy implications, budget implications, legal implications. We went through a very detailed policy review, and then at the end of that process, I said to Rod, "Hey, I need you to come in, or I'll come to your house if you won't come to the office. And we've got to go through these bills so you can decide what you want to do. Do you want to sign them or veto them?" And he kept putting me off, putting me off. And finally, I was like, "Look, Rod, the deadline is coming up on these bills. I really need some answers from you." He said, "Well, do you know what you want to do?" I said, "Well, I know what my recommendation is, sure, but I'm not the governor." He goes, "We'll just do that." "Are you sure?" He said, "Yeah." And then from there on out, we just ran the state without him. So literally, we would decide what the budget was going to be. We're talking 60, 70 billion-dollar budgets. We would write a speech, and we'd hand it to him and say, "Okay, Tuesday, you're going to read this off a teleprompter to the legislature." And he would do it, and then he'd go back to Chicago, and then we would negotiate the budget over the next few months with the legislature or the State of the State, pardons, bills, whatever it was. He just refused to do the work, and someone had to, so I did.

SPENCER: Wow, that's shocking. Is that something that you think happens fairly often, or was that just a real — ?

BRADLEY: No, he's extreme. But what was logical about his view was that, while most politicians don't say, literally, "I don't need to come to the office again for four years," when they have to decide what they're going to do on any given issue, their choice is always dictated by the next election and not what's right. Now, are there exceptions? Sure. Mike Bloomberg, who I worked for, certainly was an exception to that, but they're really very few and far between.

SPENCER: So that takes us to the last two types that you mentioned. The fourth type is ideological because it seems like at least some politicians truly believe in something. Maybe Bernie Sanders would be an example of that?

BRADLEY: Yes, yeah, absolutely. There are people on the far-left and the far-right who truly believe that we should have a socialist form of government, or that we should have a completely libertarian form of government, with no rules and regulations at all. I don't agree with that but, in a weird way, I respect them and that they at least do believe in something and they're trying to advance those beliefs, even if I think those beliefs happen to be crazy. That's the fourth. And then the fifth would be that very small group of people — like Mike Bloomberg — who genuinely just want to get things done and are willing to take political hits and do what's not in their self-interest, to try to advance something that they truly believe in.

SPENCER: There's pragmatic, basically,

BRADLEY: Yeah, are they in existence? Yeah, absolutely. But are they rare? Yes, they're unique [?].

SPENCER: Who else would you put in that category?

BRADLEY: When Obama was a state senator, I got to work with him a little bit. And actually, when I was in law school, he was a professor there. I would say that he fits in that category a bit. This is sort of a shameless plug, but I mentioned earlier that my brother-in-law is in Congress. I would say that Josh fits that mold. So there are people that do, they're are not many.

SPENCER: What about someone like AOC, Alexandria Ocasio-Cortez?

BRADLEY: She would fit into that true believer category. To her credit, I believe that she believes the things she says. I don't think she just says whatever she needs to say to get elected. I don't agree with almost anything she says. I disagree with the vast majority of it, but I at least respect the fact that she does truly believe it, and she's willing to say it even if it is unpopular. With that said, primary turnout in AOC's district is incredibly low. The people who do vote in her primary tend to be very, very far-left leaning. That's how she won her first election to begin with. And so while the position she's taking may be broadly unpopular, they're not really hurting her chances of reelection.

[promo]

SPENCER: So would you say it's fair to think that, if you're just evaluating a random politician, there's a pretty good chance that all they're really optimizing for is reelection and attention and boosting their own ego?

BRADLEY: There's a 99% chance.

SPENCER: [laughs] That's pretty high. And so what are the implications of this when we think about, okay, if you're right and we're being governed by people like this, what does that mean for society?

BRADLEY: The implications are this: the implications are school shootings because we can't, in Congress, pass any sort of real law limiting who can get their hands on an assault weapon. The implication is a massive border crisis because we are unable to put real border security into place, nor are we able to say the economy needs X million more workers to perform nursing or hospitality, construction, or whatever it is, and so we can't increase the visas and the quotas to accommodate that. It means that we have a massive affordable housing crisis in this country. It means that we have a school system that doesn't work particularly well. It means that we have a health care system that delivers the worst outcomes for far more money than any other country in the world. And just name it on and on and on. We have intractable problem after intractable problem, not because the vast majority of people don't actually agree on the solutions, but because those people don't participate in the elections that elect people. Let's take guns. 70% of voters would say we should either confiscate everyone's guns, nor should it be easy to buy an assault weapon. But that 70% don't vote in primaries; their views don't matter. 70% of voters would say we should either deport everyone here illegally, nor should it be easy to get into our country. Again, that 70%, they don't vote in primaries, so their views don't matter. So we have a tyranny of the minority, because those people, in some ways to their credit, actually bother to show up and vote in primaries, and therefore they dictate what all of our policies are. And that's why I've been funding and learning about this effort to try to make mobile voting possible. Because to me, you're never going to convince people to not take their kid to school on a random Tuesday, or be late for work, or whatever it is, to vote in a primary, especially if it's for local election, state rep, city council, or whatever it is. So you've got to meet them where they are, and we have now hit a wall where we can vote securely on our phone. That technology is being developed as we speak, that will allow people to participate in the process, not by having to go somewhere and wait in line, but while they're waiting for their coffee at Starbucks, or while they listen to a podcast, or whatever is that they're doing on a subway, they'll be able to cast their ballot, and that could radically increase primary turnout, which moves everything to the middle.

SPENCER: Now, when we talk about something like a major reform in voting, I think immediately people are suspicious, because one side tends to win and one side tends to lose. So you're going to get an extreme negative reaction from anyone who loses in that equation. We saw this with mail-in ballots. Any politician who thinks that mail-in ballots are going to disadvantage them is going to adamantly oppose it. Do you think the same thing will happen with...?

BRADLEY: [laughs] I think it's even worse, unfortunately. [laughs] I actually think that, ironically, we're going to unite members from both parties and unions and lobbyists and business groups against it because, if you know how to win elections in low-turnout primaries, you're not interested in risking that just to try to make the system work better and to make democracy better. And so we're going to have a huge uphill challenge. I've been funding and running the mobile voting project since 2017 but the reason that I wrote the book Vote With Your Phone — it comes out on September 17 — is to try to build a movement. Because the only way that we're going to win the right to make voting far easier and far more accessible is if we can bring together different coalitions of people to demand from their legislators that, "Hey, we want the ability to vote on our phone. Voting on our phone is the most secure way that we can vote, and you have to give it to us. You can't keep it from us." Unless we do that, it's not going to happen. And unless it happens, we're never going to fix the underlying problems of our democracy.

SPENCER: If you were not you, I would say, "Oh, man, that sounds impossible." [laughs] But I know that you've had a lot of experience getting things done in politics. How realistic is this and how much support do you really need to make it happen?

BRADLEY: I need a lot of support. It's very hard. The good news is, eventually we're going to vote on our phones. If I didn't exist or I never touched this thing, it would still happen eventually, because technology ultimately always wins. I don't know if that's going to be 20 years or 40 years from now, but at some point there's not going to be paper for really anything in society, let alone for voting. But I don't want to have 20 to 40 more years of a totally dysfunctional democracy when we can fix the problem much sooner. Where I really got both the idea for this and the faith in it was campaigns that I ran for Uber all over the US to legalize ridesharing. I started working with Uber in early 2011. At the time, they were just in San Francisco; they were coming to New York, and we were a tiny little tech startup, and the taxi industry was a big, muscular industry, and the way that we beat them is... We couldn't compete on lobbyists; we couldn't compete on fundraising or political donations. But we were able to say to our customers, "Look, if you want this thing to continue, if you want to be able to use this to stop your politicians from doing the corrupt bidding of the taxi industry, you need to let them know." And through the app, we made it possible to tell your city council member, your state rep, your mayor, whoever the relevant person was, "Hey, leave this thing alone. I want it," and whether that was by tweeting at them, texting them, emailing them, calling them, through technology, we were able to mobilize millions of people, and we won in every single jurisdiction in America that way. I remember when that was happening, thinking, "Oh, my God, it's not that these people are too apathetic to do anything. It's that they're just not going to show up and inconvenience themselves wildly to vote in a local primary. But it turns out, if they can do it from their phone, they will. What if we could vote this way?" So that was the underlying question that prompted the mobile voting project, and then initially we got started to answer the question. We funded elections in seven different states. We either deployed military or people with disabilities who were able to vote in real elections on their phone using the technology that then existed at the time, and it worked. Turnout: material increase. National Cyber Security Center audited all the different 21 elections and found that they all came back clean. City of Denver did a poll of the people participating in the election, and not shockingly, 100% said they preferred pushing a button to having to go somewhere. And the next phase was a lot of people in the cryptography community said, "Hey, you know what? Great, but we still don't think it's safe, and we shouldn't do this." So I realized, "Okay, you have a point. Let's make it safe." And so we built our own mobile voting technology that is end-to-end encrypted, end-to-end verifiable, Airgap as multi-factor authentication. Biometric screening is open source, so it can be fully audited by anyone. And once we are done building it, and once it gets certified by NIST — national standards and technology — we are going to make it free and open source to any governor, anyone who wants to use it. This is all being done philanthropically out of my foundation. I didn't mention, when I worked for Uber, they were such a tiny company, they couldn't afford to pay my consulting fee. And Travis Kalanick, who was then the founder and CEO, said, "Hey, would you do it for equity?" I didn't even know what equity meant at the time, but thank God, I said yes. And then Uber became this rocket ship, and became worth a lot of money. And so I've put about $20 million of my own money into mobile voting so far to try to make this thing happen. And that's a lot of the money that I made. But to me, I have both the experience in tech — I run a venture capital fund now — and in politics to really understand this issue. I didn't grow up with a lot of money. I'm a first generation American. I worked in government for most of my career, didn't make much money there. So when I got lucky and did make some money, in my view, if I have the opportunity to try to solve what is really destroying our country, I ought to try to do it. That's what we're up to.

SPENCER: I love that you're open sourcing it so that any government in the world can use it. That seems really potentially impactful if a bunch of countries can adopt it and it's actually secure. Because hacking is, of course, a very severe concern. Russia, China, there's a lot of groups that are incredibly good at hacking that could have a serious interest in messing with things.

BRADLEY: Yeah, absolutely. That's why I really want to start small. I am not proposing that we do mobile voting in this year's presidential election, or even the 2028 presidential election. I like to start local — city council races, school board races — and see how it goes, see what works, see what doesn't work, make adjustments. Because the code is gonna be open source, that means that other people can improve upon it. We'll have Bug Bounty so we can figure out what's wrong with it and catch any problems in the code and fix it and make it better. And then, over time, expand to the state elections, maybe congressional primaries, and then work our way up to every single election. But, yeah, let's start small. I just really don't see the Chinese government or the Russian government caring about a local city council election. So let's start there and see where it goes. And also, to be clear, we're not even going to put the code out there at all until we get approval from NIST. And so everything that we're doing is going to be protected by encryption, verifiable, and Airgapped. You submit your vote, you can ensure that it's accurate — exactly what you want it to be — until you cast it. The minute you cast it, it's encrypted. It hits the voting office. It's taken offline so they can't be attacked by any sort of foreign entity, printed out, tabulated, counted. So we think we have built a system that is incredibly secure. But I think if we both do that and get certified and start small, work our way up, we can build something that is truly safe, that people have confidence in.

SPENCER: And what about voter fraud? How do people actually prove that they're the person they claim?

BRADLEY: Yeah, a few things. It's up to each jurisdiction to decide how to authenticate voters. But one, we have built the system so it can do things like biometric screening: your iris, your fingerprint. In the elections that we've done with deployed military, there's both a facial recognition scan and then biometric screening to ensure that you're you, but at the very least, you're both receiving a code for multi-factor authentication — your text and your email — and then you have to submit both your salient information and, at the very least, your digital signature, which then gets mapped powerfully against the signature of yours that's on file. And then from there, you can vote. Or if the jurisdiction says you can only vote online if we determine who you are through your fingerprint, your iris, or whatever it is, even better. We built the system to be able to accommodate all of that, and so that's how we determine it. And we believe that's actually a far more secure way to identify who you are than (say) mail-in voting, which is great, but ultimately, all you've got to do is open the mailbox, take out the ballot, fill out the form and send it back in. Anybody can do that. And so we're trying to take the vulnerabilities in the current system and make it a lot more strict.

SPENCER: Right, because if it was just phone, email address, social security number, name, that information can be pulled from the dark web. Tons of this information has been stolen and released by hackers. So that's just really not enough on its own, I would say.

BRADLEY: Yeah, that's why we've got multi-factor authentication. In the same way that Citibank, in order to verify that you're you is, you're logging in and saying, "Okay, now we've sent a code to this phone number, your phone number, to this email. Prove that." And then from there, we can get into iris, fingerprint, everything else. A lot of the concerns that we hear about potential election fraud can be addressed through mobile voting, where they really can't do anything else. Although, let me also say, just because I've spent 30 years now in and around politics and elections, actual voter fraud is very, very rare. There was a study done, I think, by Tufts University. Something like 0.00006% of total ballots cast in US history have experienced any sort of fraud whatsoever. It's an issue that people like to talk about. It's great for conspiracy theories. In reality, it doesn't really occur. I've even worked in Chicago politics — that's as nasty as it gets — and even there, it doesn't actually really occur.

SPENCER: People really believe in it. Now, many Trump supporters, for instance, think that there was massive fraud in the last election.

BRADLEY: They do. I think that, as a result, it might be that mobile voting, even at the local level, happens in blue states before red states for that reason. We polled mobile voting before 2020, and about 75% of Republicans, Democrats, Independents, all said, "Hey, if all was secure, absolutely, we'd vote on our phones." We polled it after 2020, that same support stayed in the 70s for Democrats, independents, but it fell to below 50 for Republicans. And so, yeah, it may be that at least until Trump is no longer on the scene — and we don't know if that's going to be in six months or four years or God knows how long — but whenever it is, it may be that this is something that only democratic jurisdictions are pursuing; although, I really do hope that it is bipartisan. I'm an independent at this point. I don't really trust or like either party. And actually, the very first state to do mobile voting was West Virginia, and there was a very conservative Secretary of State who was Republican, Mac Warner, and the reason Mac wanted to do it is Mac served in the military; all four of his kids were in the military. And he was really offended by the fact that people who are literally risking their lives to protect our right to vote — a right that we barely even exercise — they mail their ballot from Kandahar, it shows up in the election office three weeks later after the election's over, and it gets thrown in the trash. And he wanted to stop disenfranchising the people who were putting their lives on the line to protect our right to vote. And that's how mobile voting got started.

SPENCER: It seems to me that one way to frame what you're describing here is as a behavioral nudge, where, yes, in theory, people could go out and vote in the primaries, but people are busy, they're looking after their kids, they're working and so on. So if, instead of it being an annoying thing to do, it could just be something that takes two minutes on your phone that you do at home, way, way more people would do it, because even that small amount of friction is enough to massively change human behavior.

BRADLEY: Yeah, absolutely. As an early stage venture capitalist, a lot of my time is really spent thinking about normative change. And if you reduce friction in this type of product or this type of service, or whatever it is, will that change consumer behavior? And that's exactly what we're trying to do here. A lot of what mobile voting is is a behavioral economics experiment to say, like you just said, If you remove the friction from the process and take it to something where... Basically, everybody has a smartphone at this point. According to the Pew Institute, 97% of people under the age of 50 have a smartphone. 89% of people under the age of 65 have a smartphone. 76% of people over the age of 65 have a smartphone. So we take that device we already all have — it's all in our pocket; we stare at it all day long — now all you've got to do is press a couple of buttons while you're waiting for your coffee to arrive, whatever else it is, a lot more people are going to do it, and that's how we fix the problem.

SPENCER: Now, while you're an advocate for technology and voting, my understanding is that you're actually against people betting on elections, such as through prediction markets. As some of our listeners know, there's been this rise of prediction markets, where people can actually place bets on many different things, including who's going to win political office, and it's sort of an information gathering system where you can then go look at what's the probability that Trump is going to win, or whatever politician you're interested in. So why is it that you're against that?

BRADLEY: Well, I just think that we need clear rules set around it so far. It's not that we couldn't come up with a regulatory system that works, but I think the CFTC and any other agency that is regulating this needs to be able to say — in order to have integrity in these types of bets — "We need to make sure that (for example) it is an election with X number of voters." So one thing you could worry about is, if it's a really low-turnout election, somebody could actually say, "Well, if I bet $100,000 on this thing, I could spend $50,000 on campaign flyers for a candidate and elect them and make $50,000 profit." So I think you just need to build systems and integrity into it, just like we did with the mobile voting. If we didn't have encryption and verifiability and Airgapping and everything else, then I don't think mobile voting would be worth pursuing. And I think that, if you were to have betting in elections, we need to set clear rules and regulations. And by the way, that should be true for almost every sector. For example, last week, in California, the legislature passed a bill that was the first in the country to require different types of safety testing for AI. And I think it's a shame that most of the AI companies are opposed to legislation because these are the same people that keep saying, "Oh, no, no, no, we want to make sure AI is safe. We want to make sure we're regulated." And then when actual responsible regulation comes their way, all of a sudden, they're against it. You can't live in a world in the absence of all rules and standards. By definition, human beings are successful because we have structure. You could go way back to the beginning of humanity; you had Neanderthals, you had Homo sapiens. And the reason why the Homo sapiens won even though the Neanderthals were bigger and stronger and smarter is that they couldn't work together; they couldn't cooperate. Homo sapiens could, and that led to structure; that led to governments, led to systems. That doesn't mean that every system is good or every rule is good, but the absence of all rules is also not good. And this is a big problem that I have, both with the tech industry and a lot of people on the Left in the government or policy side, which is, a lot of my colleagues in venture capital will say, "Oh, all regulation is bad and we should be libertarian." That's not true. And then a lot of people that I know and work with on the policy side who are running agencies and things like that, will say, "Oh, all regulation is good and all business is bad, and we can't trust the markets at all." And that's not true either. A regulation is as good or bad as the context that it's applied to, and we've got to think about it intelligently that way, instead of taking these blanket positions that really don't make any sense.

SPENCER: I find your view on regulation a little surprising just because you've spent so much of your career basically trying to figure out how to work around regulations to get things done.

BRADLEY: Yeah, it just depends. There are times where regulations can be really stupid and there are times where regulations can be really necessary. I would say that, as a VC, my work falls into one of two buckets. Either I invest in a company (say) like Uber... Uber didn't invent the notion of paying someone to take them from point A to point B; they came up with a better way to do it. And because we came up with a better way to do it, we attracted a lot of customers. That took market share away from the taxi industry, and then they started using political pressure to try to shut us down. That was corrupt. Those were regulations that existed solely to prevent competition, not to enhance consumer safety or anything else. They were fraudulent regulations, and we defeated them. But then we also invest in total white spaces like AI or crypto or machine learning or drones or autonomous vehicles, where there's no regulatory framework whatsoever, and you can't build industries and multi-billion dollar companies in the absence of all structure, the absence of all rules and regulations. In those cases, the work is to create the right regulatory framework. And that doesn't mean that they should be overly burdensome, but it's to figure out what the rules should be. Where is the jurisdiction? Should it be municipal? Should it be state? Should it be federal? Who should it be? How should it work? And that's really important. Take crypto. I was an investor in Coinbase; we're investors in Circle. You can't have legitimate crypto exchanges in the absence of all rules because, otherwise, you're gonna have all these fraudulent ICOs and all these fraudulent exchanges. And if people don't have any confidence in the trust and integrity of the system, they're not going to do it. So if you're Coinbase or if you're Circle, you don't want unreasonable regulations, but you don't want the absence of all regulations either. Same coin is true on betting markets for elections or anything else. Again, it's not that regulations are inherently good or bad; it's that they have to be thoughtful and reasonable to promote innovation but also protect the consumers.

SPENCER: I think crypto is such an interesting example because, in many ways, it works around regulation, where you can just go do a thing and it can be running on the blockchain and nobody can even shut it down. While that's really appealing in some ways, we also see how it can be so chaotic where people accidentally put in the wrong address and their money's just gone forever. Or people have their account hacked and they lose their money all the time and there's just no recourse for them. I think it both shows the potential for doing things outside the system, but also the incredible cost of having no regulatory framework.

BRADLEY: Yeah, absolutely. I think that crypto is an interesting use case because, by definition, the true believers will say, "We don't trust central banks. We don't trust central governments," who would rather trust our law with strangers whose names we'll never know over the blockchain because they are ideologically aligned with how we see the world. And that's, I think, totally fine; there may be a smart hedge against various types of geopolitical risk. At the same time, you also are at the risk that, if you lose the encryption, then you're not able to recover that money at all, or it could be stolen from you or something else. And so, yeah, the people who choose to engage in it. But nonetheless, when there are crypto exchanges and Initial Coin Offerings happening, there's no reason that it shouldn't comport with basic security rules to give consumers and investors a certain amount of protection as well. And it really gets partly the question of, is cryptocurrency really currency, or is it a security? If it were truly a currency, then it would work by one set of rules, security by another set of rules. There's not really much you could actually buy with crypto as a currency. It's mainly used as an asset class for people to trade and try to make money or hedge other bets or things like that. And so part of it depends on where it evolves into. If, one day, we are truly buying business services with Bitcoin or Ethereum or whatever it is, that would argue for one sort of regulatory structure, and if it's purely just another way that we choose to deploy capital, that's done [?].

SPENCER: Going meta for a second, when people think about regulation, they'll have some people that want to regulate everything and not let the free market do its thing. And you have other people on the side saying, "No, no, no, regulation is always bad. Just let people do exactly what they want to do all the time." It just seems to me, regulation is an incredibly powerful tool, and like any tool, it can be used effectively or it can be abused. And like you said, it's really about in which specific instances is it actually helping and in which is it harming? And if you think about just standard economic theory, it actually provides quite an interesting framework for this to tell us when regulation is going to be more helpful than harmful. In particular, standard economic theory says that, if you have information asymmetries, that's one place where the market will fail. If you have monopoly players, that's one situation where the market will fail. But if you have situations where everyone knows what's happening, and there's low transaction costs, and people have a sense of what they actually want, then actually it can be incredibly powerful just to let people transact however they choose. I think we actually know quite a bit about the situations where regulation will be more helpful than harmful, and also the reverse.

BRADLEY: Yeah, absolutely. Literally, my job title is venture capitalist so I am a capitalist through and through. But markets are not absolute and they're not perfect. You certainly want free markets. Capitalism, for whatever flaws it has, is still the greatest force or economic good in progress in the history of the world. There are vastly fewer people now living in extreme poverty. There's much less infant mortality, much longer life expectancy, much more literacy. All of that as a result of countries all over the world embracing some form of capitalism, and the markets bringing wealth and opportunity to people to make their lives better. Capitalism, in many ways, has been the greatest force for good in history. At the same time, it only works if there's a certain set of symmetry and information and fairness in the system, and that does require regulation.

SPENCER: I think another flaw with free market capitalism is that it tends to lead to accumulations of wealth. If you have enough money to make an investment, you can earn at an exponential rate of return on that investment; whereas, if someone is living paycheck to paycheck, they literally can't make an investment at all. So there does seem to be certain weaknesses that way. And then you start getting into this idea, well, maybe capitalism needs redistribution to make it work well.

BRADLEY: Yes, although I don't know that capitalism and redistribution are mutually exclusive. For example, I would argue that we want a system that allows the total greatest amount of money by society to be made; that is, a wealthier, more prosperous society with a higher quality of life than otherwise. But it also results in very unequal distribution of wealth, and income inequality, and then the question is, what's the right way to redistribute that money? I would argue it's universal basic income. Having spent time running government budgets, while, in theory, you can redistribute wealth from the wealthier to the poorer through taxation leading to different government programs, there's a lot of slippage along the way. One dollar of your taxes, by the time it reaches a person in need, by the time that you're done with pork and appropriations and union contracts and bureaucracy and everything else, maybe 50 cents is actually going to help the person who you would try to help with that dollar. One of the things that I really like about universal basic income is a dollar of wealth is a dollar of wealth. And if it goes from my pocket to your pocket, you now have a dollar to spend to feed your family, to pay your rent, to save up to send your kid to school, whatever it is. I do believe that there are certain functions of government — infrastructure, public safety operations — that are essential, and you're always going to need taxation to fund those services. But I also think that a lot of social services that we do will ultimately help a lot more people if the money was sent directly from point A to point B, as opposed to so much getting skimmed off the top.

SPENCER: Yeah, I think it's a really interesting question of when giving money directly is superior to giving services because there are times when a service can be more valuable than money, such as, suppose people in an area are dying of a disease and there is no treatment available for it, being able to provide that treatment, okay, maybe that really helps them. Or maybe if someone is severely addicted to a very dangerous drug, maybe there are times when you really just need to give them a service to help them get out of that bad spiral. On the other hand, as you point out, I think a lot of services are just extremely inefficient and maybe they're just not even helping people very much, and so people might consume them, but that doesn't mean they're really benefiting.

BRADLEY: Right, or if there are economies of scale. The other thing that we actually do out of my foundation is school meals. Legislative campaigns in states demanding programs like universal school breakfast, universal school lunch, those work incredibly well because schools are able to provide those meals to kids for 70 cents a meal, and so you can feed a tremendous number of kids for not very much money. In that case, the economies of scale work really well. But there are other cases where we try to provide services that are so tailor-made and need to be so targeted that it would be procured a lot more cheaply and effectively through the private sector. So it really depends. I think that we should be able to look at every single program and audit it and say, "Is this working or not? What's the ROI for taxpayers here?" And if the answer is, "It's a good program because a bunch of special interests are getting money off of it, but it's not really helping people," then we ought to do it differently. But again, we're never going to get to that level of thinking and independence and integrity in government unless we make the special interests a lot less powerful, and that means a much higher primary return. It also gets back to mobile voting.

SPENCER: Another type of special interest that a lot of people are concerned about are large tech monopolies, and there's an interesting debate going on there. Some people say, "Well, look, large tech monopolies today, they're not really hurting consumers. People use Google because they like Google. They're not really forced to use Google. They could use alternatives. They just like it better." Others say, "No, that kind of concentration of power is inherently bad for society. It reduces competition." Where do you stand on giant tech monopolies, and do you think that they should be broken up more?

BRADLEY: I do, I do. I am actually a very big supporter of Lina Khan's efforts to try to break up big tech and to better regulate it. I've actually been very active with Lina and working with her on this. I've been out there publicly in support of it. I happen to own a bookstore on the Lower East Side, and she and I did an event there together. She came on my podcast. I really do believe that, if we want to encourage innovation and early stage company formation, we've got to go after big tech. I am an early stage tech investor. I invest in Seed in Series A so it's really early in a company's life cycle, and there is no way that I could invest in the potential competitor to Meta or Apple or Amazon or Microsoft, because they have monopolistic power and they can just crush anyone in their way. The problem is, while those companies seem like these amazing innovators today, every single company eventually grows stagnant. If that weren't true, Standard Oil and General Electric and IBM and Union Carbide or whatever, would still be the biggest companies today, and they're not, simply because reality always sets in, and the founder who had an incredible vision eventually dies or moves on, and other people come in, and their priorities change and they become more bureaucratic. So some of those companies are going to stagnate over the next ten, 20 years, and if no new companies are being formed today to try to compete, then when those companies stagnate, there's gonna be nothing to take their place. And you're gonna have a giant hole in the economy and a giant loss of jobs and wealth and tax revenue and consumer services and products. So the only way to protect the economy long-term is to ensure that these companies don't have monopolies. I think a lot of the efforts by the FTC and DOJ to prosecute Google and Meta and Amazon and Apple and Microsoft and others make a lot of sense. At the same time, that doesn't mean that all corporate M&A activity is necessarily bad. While I agree significantly with Lina in her antitrust policies around big tech, I would say that her general activity towards mergers and acquisitions of corporate activity has had this massive chilling effect on M&A, which has led to a lot less investment in new businesses and innovation. As a venture capitalist, 90% of our exits come through M&A deals, not through IPOs. When there's no liquidity in the marketplace at all because every single merger is being protected by the FTC, then ultimately companies can't sell. Venture capitalists aren't getting capital to return to their investors or put into new deals. Far less money is being invested. It's far less new company formation, far less innovation, and that would really hurt the economy. And so it gets back to that discussion we were having a few minutes ago about regulation, which is, even within the world of antitrust, it's not inherently good nor evil. It really depends on the application. The application of antitrust policy against big tech, I think, promotes innovation. The application of antitrust policy against all M&A really hurts innovation. So it's a matter of being thoughtful about understanding what's what.

[promo]

SPENCER: It sounds like your concern about these big tech monopolies is not so much their behavior today as it is that you're worried they'll stagnate in the future? Is that right?

BRADLEY: It's both. From an economic standpoint, it's about future stagnation in the nation. I would say, from a social policy standpoint, I am also a strong supporter of repealing section 230. Do you know what that is or should I explain that for the listeners?

SPENCER: Could you explain that?

BRADLEY: Sure. Section 230 was a little provision — 26 words that were tucked into the Communications Decency Act in 1996 — and what it says is that internet service providers cannot be sued for content posted on these platforms by its users, which actually made a lot of sense back then, because for the Internet to take off, there had to be some freedom from liability for this company to develop. What no one could have foreseen in 1996 were social media. And now we are in a world where Meta, Instagram, Facebook, Whatsapp (from Meta), or Tiktok or YouTube or Twitter or any of these platforms, are completely immune to any liability. So if I defame you on Twitter, you can sue me. But where the real deep pockets are is Twitter itself, and they are completely immune to any liability whatsoever. And the problem is human beings inherently having negativity bias; that's an evolutionary trait that we've developed to keep us from being eaten by lions way back when, or now when we smell gas, in order to leave the house. You have to have a negativity bias in order to survive the world. Where it also is that, when we see negative headlines and negative clicks, we are far more likely to click on that than we are positive content. And guess who knows that? Elon Musk, Mark Zuckerberg, the people at YouTube, all the different platforms. They have a perverse incentive to promote the most negative, toxic content possible, because that generates more clicks, which is how they generate more advertising revenue, which is how they generate more profits. And so they are incentivized to push the worst of the worst content out towards us. And what has that led to? Massive mental health crisis, massive increase in depression, teenage suicide, self-harm, sex trafficking, guns, all these terrible things, simply because we protect these companies from any sort of liability and that has led to all these terrible outcomes. For example, I don't know if you read the World Happiness Report last year, but the United States, which is the wealthiest country in the history of the world, when it came to happiness for people under the age of 30, we're 68. Think about it. We have more money, more resources, more opportunity than anywhere else ever, and yet our young people are miserable. Why are they miserable? Because social media is making them miserable. If you think about it, what is social media? It's the unhappiness machine. It does two things automatically. You go on there and you see: (a) "Here's how my real life compares to everyone else's fiction," and you feel inadequate immediately; and then, (b) everything bad happening anywhere in the world's all of a sudden apparent to you all at once, all of the time. Your life feels inadequate and the world feels terrible, and that makes you unhappy. So social media is having this incredibly destructive impact on society, and yet we refuse to regulate it, and that is having a terrible effect on our kids. So I am a staunch advocate for having Congress repeal Section 230, take away this liability protection, make the companies moderate content, make them maybe be more socially responsible, and I think, we will all be much better in the long run.

SPENCER: I'm confused how that would work in practice. Imagine that someone called someone out for something that they think is bad, but they're wrong about that; it's false information. So then the social media company could be sued because someone made a false accusation? I don't see how that could function.

BRADLEY: Yeah, it would probably be less around me insulting you. I don't think that Facebook would be liable for that. But if I'm putting out...

SPENCER: Well, defamation. Sorry, I don't just mean insulting but defamation.

BRADLEY: Right, but if I am putting out false claims to try to influence an election, or to tell kids that using — I have a 15-year-old son; luckily, he doesn't seem to do this — but Zyn is one of these tobacco alternatives, things that they pop in their mouth, is cool, and they all know about it from Tiktok or whatever it is. When there are societally harmful behaviors that the algorithms are pushing, that's when the companies become liable, not unlike the tobacco litigation in the 1980s. Smoking was legal — it's still legal — but when tobacco companies are pushing cigarettes on minors and other protected classes, that's when there was liability, and the fact that they were liable and have these multi-, multi-billion dollar judgments against them is what led to real reform changes and helped lead to the reduction in cigarette smoking in this country. So, yeah, basic defamation, sure, that's probably not going to change. But when the platforms are knowingly allowing really harmful content — teaching teenage girls, here's how you cut yourself, here's how you have an eating disorder — and they are knowingly letting that go forward because they know it will generate more clicks, that's when they're liable, and that's what you can change.

SPENCER: It seems just really tricky to me, because obviously there's really harmful content on social media, and it would be great if it was removed. But on the other hand, I don't see how social media could ban all harmful content without essentially just having a massive censorship program where lots and lots of acceptable content gets banned, too.

BRADLEY: Yeah, you would certainly have some risk of being over-inclusive, and I get that. But ultimately, what would happen is, if you remove the liability protection, over the period of years, a series of jurisprudence will emerge in the courts that will determine, here's how to do it, here's what works, here's what doesn't work. And there's not going to be a one-size-fits-all situation. But ultimately, if over a 10-year period, people are going to court and saying, "Hey, YouTube really shouldn't allow those ISIS recruitment videos that my kid that ended up..." Not specifically, but there actually has been litigation on this, that someone's kid that ended up being recruited by ISIS and becoming a terrorist, that a court will say, "Yeah, Google should have blocked that, Google and YouTube," or "No, Google wasn't liable for that." And so the jurisprudence will emerge over time, and that will set the rules. It's not going to be simple, but when the alternative is an internet that is just filled with harmful content and toxicity, you've got to do something.

SPENCER: Yeah, it's really interesting. I personally don't know which way is better, because it could be that the equilibrium is something reasonable, or it could be the equilibrium is all the platforms are just massively censored, and there's tons of very reasonable opinions you can't express on any social media now.

BRADLEY: Yeah, except the First Amendment's not going away in this country.. The platforms are going to claim the First Amendment in every case, and the plaintiffs are going to say, "Yeah, but..." just like if you yell, "Fire!" in a crowded theater, if you are knowingly promoting content about bulimia that's positive to 13-year-old girls, that expands beyond your right to free speech, and that's where you're liable and that's where you should have moderated the content, and that's where courts are going to come in. And sometimes you're going to have reversals when you go to the Supreme Court, and different options are going to get set. It's not going to be simple, but I do think that, over a period of time, you can develop a doctrine that is thoughtful.

SPENCER: Last topic before we wrap up, I know that you have opinions about how AI could be used to help with government regulation and make things more effective. Where do you see AI fitting in there?

BRADLEY: I think there's a lot of opportunity. As you know — because you're a New Yorker like me — the New York rollout of legal cannabis has been an utter disaster. So 2021, the Albany...

SPENCER: It's so bizarre.

BRADLEY: It's crazy. So Albany legalizes cannabis in 2021 and they set out, "Hey, here's how we should go about licensing dispensaries," and they create something called the Office of Cannabis Management, which is just a debacle. It takes them two and a half years to promulgate regulations and, in that period, while they're basically just twiddling their thumbs, thousands and thousands of legal weed shops pop up all over the city. They're literally, as you know, every single block, and New Yorkers, it's hard to really expect your average person to know like, "Oh, is this licensed? Is it not licensed?" They know that Albany said it was okay and...

SPENCER: But they're basically all legal, all the initial ones?

BRADLEY: All legal. So there are 140 licensed dispensaries today in New York, and over 5000 illegal dispensaries. And the city and the state both pointed the finger at the other and said, "No, it's not my responsibility to close these; it's yours." And so, for years, these dispensaries were allowed to thrive. It ended up creating situations where teenagers across the city were able to get their hands on weed constantly, creating a new addiction crisis, and I wrote a piece on my Substack arguing that, had we used AI instead to both promulgate the regulations and then determine who does and doesn't get licensed, as determined by the criteria set by the legislature, we could have worked this all out in a matter of months, not years, and instead of having this vacuum where an illegal market popped up, you could have had a sufficient number of licensed dispensaries to meet consumer demand, to comply with all the short products were inspected and regulated, ensure that people above the age of 21 were getting them. You could have used AI to achieve a far different outcome, far superior here. Whether it's for things like licensure or procurement, there's so much corruption in government procurement, where people get paid off to steer a contract one way or another, or there's massive waste — you pay $1,000 for toilet seats or whatever it is — all of that could be done so much better through AI. AI has plenty of risk, plenty of potential harms. There should absolutely be regulation of AI. There should be safe testing for AI. But at the same time, there are going to be opportunities to utilize AI to make government work better for people, and we should take those out.

SPENCER: What would that look like in practice for the cannabis industry? I'm having trouble visualizing what AI would be doing.

BRADLEY: What it would look like in practice is that the Office of Cannabis Management, instead of having all of these employees who would take three days to just set up meetings with each other, basically say, "Okay, this is the AI system we're going to use. These are the inputs that we're going to use. The legislature said that these are the priorities. Let's promulgate a set of rules that (say) the Governor's Council can review to take all those priorities into account" — take all the different issues into account, of things that could go wrong, or you thought about, or compliance with different statutes or whatever else — and produce and promulgate a set of regulations that could then be improved and issued very quickly, instead of taking years and years and years. And by the time they finally did do it, the court then overturned it because they couldn't even comply with the laws then. Or imagine that, for licensing, rather than having these committees and people meeting — nine different business units within the Office of Cannabis Management reviewing every single application — you say, "Okay, we know we want to give precedence to women-owned businesses, minority-owned businesses, veteran-owned businesses," whatever it might be. We want to have no more than this many dispensaries for this many square blocks. We want licensees to have these types of experience and criteria, this type of capital requirements. We want people to see that their lease is signed, whatever they are. Based on all that, that's how the application is constructed. People submit them. They're reviewed by the AI, and the AI quickly spits out recommendations that, "Okay, based on all these criteria, of the 5000 applications we got, these are the top 200 and these are the ones that should be licensed." And then again, you could have a human being review that process if you want, but you can do it in a fraction of the time.

SPENCER: It's funny, with that example, where essentially you're talking about building a screening system that would recommend which people should get the license, I feel like even a simple set of rules, a point scoring system, where whatever criteria they're interested in, you get a point for each of the criteria you meet, that obviously might work better than... [laughs]

BRADLEY: Well, they have no problem with that. That is what they did. It unfortunately didn't work simply because what you're saying from a theoretical standpoint, it literally took them years just to have enough meetings where all the different stakeholders and all the different groups and everything else to determine the points. The final determinations might have been reasonably accurate or reasonably similar to what the AI would have come up with, but what AI could have done in a matter of days or weeks, people were taking years and years to do instead, because they need lunch breaks and they need vacations, and Martha's out on Tuesday, and Dan's got to drop his kid off at school on Thursday, and all of a sudden, weeks and months go by with no sense of urgency. Licensed dispensaries are not happening. The illegal market is sprouting up and filling the hole, the vacuum. All of a sudden, illegal weed is available everywhere. And by the way, not only is it leading to teenage addiction; it created, in my view, this culture of lawlessness in New York City where, if illegal drugs can be procured openly, it sets a tenor that anything goes. Now all of a sudden, people are riding bikes on the sidewalk, they're riding them the wrong way down the street and urinating on the sidewalk. We've got scaffolding everywhere, got mentally ill people shooting up fentanyl and harassing people on the subways. And New York City went from a lovely place to live, to a place that is much harder — quality of life has gone down significantly — and I think a lot of that was the result of the failure, ultimately, to control the weed crisis.

SPENCER: From my perspective, I wonder whether, if we were to use AI for something like this, would we end up just pushing the debate, instead of being about the point system to being about what prompt do you put in the AI or what are the constraints of the AI? And I just wonder, is that really an easier problem to solve, or is it that now, we're debating that instead? Do you know what I mean?

BRADLEY: At the very least, you're now debating inputs that could at least then get you answers quickly, as opposed to over an incredibly long period of time. Yeah, there's always going to be some sort of input that is going to be subjective in some way. And by the way, that's the point of the democratic process. You know what should not be determined by AI? Legislation should be the product of different people with different points of view coming together and having the ability to work with the other side, because they're not bound by tiny primary turnout, where only the base matters, and saying, "Okay, in the process of compromise and consensus, we believe that the law ensures A, B, C, D, and E, and those should be the inputs." That's democracy. And I think that we can combine the human elements of democracy, where legislators are actually able to work together, because they are not controlled by a low-turnout primary system, with the advantages of AI that we're then allowed to take those inputs and run them through the system a lot faster to reach outcomes. That, as a combination, will be better. Perfect? No, we're never going to have perfection in anything that we do, but we should certainly strive for improvement and this could get us there.

SPENCER: One final question I'm just curious about before we finish up. You worked with Uber trying to get them to essentially be legal. I'm wondering, what did you find to be the biggest challenge there, when you were pushing to get the regulators to accept Uber?

BRADLEY: The political influence of the taxi industry. The taxi industry was terrible at servicing their customers, terrible at treating their drivers, terrible at providing good, safe, clean vehicles, but they were great at hiring lobbyists. They were great at giving out campaign contributions. They were great at building political muscle and flexing it. Basically, everything we were up against was not legitimate policy disagreement. It was political corruption. And it was just so widespread that we had to ultimately manage to organize billions of our customers to be heard, to push back on that successfully. There are just far too many examples across our system where special interests don't have a valid policy argument, but they have political muscle. They have campaign contributions, they have lobbyists, and they use the power of those things to prevent good things from happening for consumers. And that, in my view, is wrong.

SPENCER: It seems like, if you have a system set up a certain way, people will find some strategy that gives them power within that system, and then they'll resist any kind of innovation that's going to dislodge their power. Some have accused real estate brokers from doing this as their job became less essential due to online search, yeah.

BRADLEY: I've been leading the fight in the New York City Council from the private sector side, to actually ban brokers' fees for rental apartments in New York City, because to me, it is totally a monopolistic sort of corrupt practice where people — who are just trying to rent an apartment and all the information they need is readily available on the internet — have to pay 15% of their annual rent to a broker for literally nothing? That is theft. And we've got legislation in the city council that would actually finally put a stop to this. We've got 33 co-sponsors on the bill. We need 26 to pass the bill. If the mayor vetoes it — which I suspect he might, because he's very influenced by the real estate industry — we need 34 to override now. But yeah, absolutely.

SPENCER: Bradley, thanks so much for coming on.

BRADLEY: Spencer, thanks for having me, man. This was such a fun conversation.

[outro]

JOSH: A listener asks: "To what extent do you think conscientiousness and grit are malleable? How should someone change their perception of themselves, for example, if they're unable to achieve their goals in life but suspect it's due to low conscientiousness?"

SPENCER: I do think that personality is fairly fixed in adulthood, meaning that most of the time if you measure someone's personality today and then you measure it again in a year or five years, it will be pretty similar, quite similar. That doesn't mean though that you can't change aspects of yourself. First of all, some aspects of personality can change. For example, neuroticism, which can involve anxiety and depression, [and] anger, there are treatments that can help people with that and they can end up less anxious, less depressed, less angry, right? So, I wouldn't think of it as permanently fixed even though it often doesn't change very much, right? So, there's a difference between on average it doesn't change versus you can't make it change. There's also a lot of behavioral stuff that's not really personality, right? If you think you're low conscientiousness, okay, are there ways that you can be more like a conscientious person even if you're not going to change your personality? Can you use systems that help you? Can you, for example, create a habit where every morning you do exactly the same thing and then you can put stuff in that system to make sure you get it done and every time you have something to do or something to remember, you record it in a sheet of paper and then you know it's always in one place? So, my point is that you can kind of hack conscientiousness to a significant extent. It's not the same as having it, but maybe it can compensate a bunch. So, the first thing is maybe you can change your personality to some extent. The second thing is maybe you can get around it. Maybe you can find ways you don't need to. And third, I think it also depends on what you're trying to do in life. There's some jobs where a certain personality trait will really, really benefit you, and if you don't have it, you'll be at a huge disadvantage. But maybe there's other things, other paths in life where you don't need that as much. Some jobs are more conscientiousness-loaded than another, so that's worth keeping in mind as well.