Odd Lots - Meet the Politician the AI Industry Is Trying to Stop

Episode Date: December 18, 2025

The politics of AI are already exploding. Whether we're talking about data centers, electricity prices, labor displacement, water consumption, competition with China, or users of chatbots becoming psychotically obsessed, AI is already a major topic in elections. And since there's so much money at stake, the industry is already spinning up super PACs and lobbying arms. Last month, it was reported that a new $100 million AI-industry super PAC called Leading the Future would be directly targeting Alex Bores, a Democrat who is running for his party's nomination for New York's 12th congressional district. Why target Bores? Well, as an New York assemblymember, he has led the push for the regulation of AI at the state level. The industry, of course, views state-level regulation as an existential threat to their business. So on this episode we speak with Alex about how he views AI and the optimal approach to regulation. Alex also has a tech background, and so we talk about the technology more broadly, as well as other issues in contemporary politics.Subscribe to the Odd Lots NewsletterJoin the conversation: discord.gg/oddlotsSee omnystudio.com/listener for privacy information.

The news doesn't stop on the weekends. Context changes constantly. And now Bloomberg is the place to stay on top of it all. Hi, I'm David Gurra. Join us every Saturday and Sunday for the new Bloomberg this weekend. I'm Christina Rafini. We'll bring you the latest headlines, in-depth analysis, and big interviews. All the stories that hit home on your days off.

And I'm Lisa Mateo. Watch and listen to Bloomberg this weekend for thoughtful, enlightening conversations about business, lifestyle, people, and culture. On Saturday mornings, we put the past week's events into context, examining what happened in the markets and the world. That on Sundays, we speak with journalists, columnists, and key political figures to prepare you for the week ahead. Join us as soon as you wake up and bring us with you wherever your weekend plans take you. Watch us on Bloomberg Television. Listen on Bloomberg Radio, stream the show live on the Bloomberg business app, or listen to the podcast. That's Bloomberg this weekend.

Saturdays and Sundays starting at 7 a.m. Eastern. Make us part of your weekend routine on Bloomberg television, radio, and wherever you get your podcasts. Bloomberg Audio Studios, Podcasts, Radio News. Hello and welcome to another episode of the Odd Lots podcast. I'm Joe Wisentham. And I'm Tracy Allaway. You know, obviously we've been talking about, I think 2028. It's going to be a big election for AI.

Really, 2026. Actually now, you know, it's not really even a prediction. This is just a description of fact. AI is going to be very big for politics. Yeah, I think it's inescapable at this point. AI is sort of dominating the news cycle as well. I know it's a running joke on the podcast, but every time we record an AI related episode, another headline hits the terminal about, you know, some new billion dollar billions. And what's the one that we just got? Disney to make one billion

dollar equity investment in open AI. Literally every time we do an episode, especially about AI, but even though there's some headline about a new investment, which just goes to show. But you know, Well, the fact that it's going to dominate politics is the least surprising thing ever because it touches on everything. Anything that's politically sensitive, the labor market, right? That's the obvious big one. Electricity cost, water consumption, wealth and inequality, and who has the power and who's accumulating more and more money in the tech industry and who's not and all this thing. Like, there is hardly a political topic that in some way, I feel like AI does not exacerbate in some way. Absolutely.

Meanwhile, the other big news politically when it comes to AI this week is that Trump issued an executive order for a national rule on AI. Yeah. Which a lot of people who are trying to safeguard the system and protect data, privacy rights, that sort of thing, do not want. Well, this is the other thing. So, there's all this anxiety for various reasons. Pick your poison, whether it's going to displace the jobs, whether AI is going to be too woke and come up. with versions of history that people don't like, whether it's going to, you know, just rot our minds with slop, or, you know, whether it's going to turn us all into paperclips. So you say,

okay, we should regulate. But what does that even look like? Like, what is a sort of productive regulation look like such that, okay, hopefully we could maintain positive aspects of the technological development while capping the downsides? Right. That would be nice. I would love that, but, like, how exactly do you do that? The argument that you hear from AI. proponents is that any regulation needs to balance, you know, safety with innovation because there's also the question of China, which that's another hot button political topic, right? This idea that the U.S. is in an existential battle with China over AI and we have to win at all costs. Therefore, the industry must not be tightly regulated. Yeah, right. So it's also a geopolitical

element. Anyway, so we mentioned AI is going to be in politics. We really do have the perfect guest because we're going to be speaking to someone that the AI industry is actually targeting someone that the AI industry is actively trying to stop. There's this new super PAC, and I think it has some entries and money and maybe a little some executives from open AI, etc. Anyway, they want to like, you know, make sure that sort of AI sympathetic politicians do well, but they're also targeting politicians that they perceive to be too negative and they've even named names, which I think is super interesting, and we have them. So we are going to be speaking with Alex Boris. He's a state assembly member here in New York. He's also a candidate in the

primary for the 12th district, which is here in Manhattan, which I feel like a thousand people are going to be running for this open seat. I feel like if you were targeted by the AI industry, that is probably some of the best advertiser you could get in a very crowded field where it's just like there's going to be a million choices. Also, yeah. I was going to say what you're going to say next, I think, but he used to work at Palantir as a data scientist. And Palantir is one of the companies backing that super pack. So you know what I think we should do? Can we talk about AI for like five minutes and then make this a 50-minute episode about

what Palantir does for someone who used to work? That's my goal for this episode. Why don't we just make it a series? Let's make it a series. Anyway, Alex Boris, thank you so much for coming on odd laws. Thank you for having me. So what's the deal like people are flying in from around the country just to run for this 12th district seat that's open up?

Yeah. Most recently, George Conway, who lives in Bethesda, Maryland, said that he was going to move to the district and run. You just have to live in the state. You don't have to live in the district to run. So last week, someone who lives in the Bronx also declared for this. Is this the one? Jack Schlausberg. Is he running for this one? Jack is also running. Yes. I think there's 10 declared candidates now and more to come, I'm sure. What a farce. How did you actually transition from being a data scientist at Palantir to politics? It's not a natural progression for a lot of people. Well, what brought me to Palantir in the first place was the ability to work with

government and actually. What are we talking about by the way? So I joined Palanty. I joined I left in 2014. I left in 2019. And so many people think of government as just passing the bills, but that's not how most citizens interact with government. It's can I get to the DMV and get my license renewed quickly? It's starting a business. It's the day-to-day interactions. And that implementation was an exciting thing to work on. So I was at Palantir for four and a half years. I then went to a couple of startups afterwards, one that used early transformers, Burt and Laser, to help banks with anti-money laundering and counter-terrorist financing. and then a startup that worked with municipalities to better distribute aid. So I had this through line of actually having government deliver on its promises throughout. We're going to talk more about Palantir later. So I'm reading this is from there's a recent political pro article. AI SuperPEC leading the future say they're spending up a multi-million dollar effort to sink Boris's nation primary campaign to replace retiring Manhattan representative Jerry Netler. Okay.

So what is it about your time the State Assembly such that you got on their radar? Most prominently, I've done a number of bills while there have passed 27 bills in my three years, which coincidentally is the same number Congress as a whole passed in 2023. But there's one bill in particular that caught their eye, which is the Rays Act. And this bill would for the first time put safety standards on advanced AI research. They really didn't like that bill. And so they announced me as public enemy number one. And the initial announcement said they're planning to spend multiple millions against me.

Last week, they upped it to $10 million. I'm hoping if the campaign continues, I can use up all $100 million that they've plans. But we'll see where it goes. What was in the Rays Act? Like, what was the gist of this bill that they hanged? Yeah, so it was requiring the major frontier lab. So we're talking meta, Google, XAI, OpenAI, Anthropic. That's probably all it would apply to right now.

I think over time maybe Amazon gets in there. deep seek and maybe mistral, but it's a single-digit number of companies. And they would have to have a public safety plan that they disclose and stuck to. Also disclose critical safety incidents, so things that go wrong that could lead to increased risk, your model weights get stolen, you lose control of the model, et cetera. And that if your models fail your own tests, that you can't release that model. And that's designed to counteract what we saw with the tobacco companies where they knew that cigarettes were causing cancer, but denied it publicly and to release the product. This is saying, hey, you companies, you're the experts. But if you're

getting reports back that this is very dangerous, you need to take action on that. It also proposes fines for violations, right? I think it's up to $10 million for the first, and then subsequent ones are $30 million. Does that actually matter to tech companies? Like, I know they care. They care about money, but like if Google is pulling in $100 billion per year, it can, you know, it can pay $3 billion fines? Absolutely. In my opinion, those fines are too low. The original version had 10% of their training costs and it would scale up with what they're investing. But generally in New York, we don't like an uncapped maximum fine. So that was part of the negotiations. I agree. I think there could be companies that just say, we're going to ignore this. I think the fact that they're spending

nearly that amount lobbying against it suggests that they do think it has some teeth. Talk to us a little bit more about what specifically would trigger the fines? Because I could see some very perverse things going on. First of all, I could see smaller companies that are right at the edge where it's like, okay, maybe this isn't a big deal for Google, but this is a bigger deal for a smaller company that's just starting out. We don't want to lock in just the handful of biggest players, right? Because regulation can serve that. It can perversely lock in the incumbents because they're the other ones who can deal with the regulatory thicket. There are also issues that could arise where, well, if I'm going to get fined really badly, I'm just, I'm going to look the other way. I'm not going to notice what these safety violations are. But talk to us about what are, in your view, the formal triggers such that the bill does not have these sort of perverse incentives.

Yeah, I feel like this is the form where I can really get into the details on it. So let's dive in. There's a two-part test for the Rays Act to apply. The first is that you're a company that has spent $100 million specifically on compute, specifically on the final training run of models. So not the tests that lead up to it, but actually the final pre-training and post-training before you put the model out.

That's one part. So it doesn't apply if you're an academic institution. It doesn't apply if you spent less than $100 million. The second part is how do you define a frontier model? And the easy way is you have trained a model that by itself was $100 million in compute and had 10 to the 26 flops, computational operations in training, It's a measure of the complexity and the size of the model.

That's a standard one that people have dealt with for a while. And so that's when I listed the five companies that I think it applies to. Those are the companies I think have trained one of those $100 million, $10,000 to the 26. But there's a second way that you can be a frontier model, and that is you are trained via the specific process of knowledge distillation, and at least $5 million is spent on that. So knowledge distillation is a technique of training a small model based on the outputs of it. a larger model. And it's a way to kind of shortcut a lot of the training and get a smaller model that has similar capabilities. Importantly, it's the technique that China has been using to

catch up to the U.S. because of our limits on compute to China, because of the export controls, that's how they're making their advanced models. And so this is the only bill that I know of that would apply some regulatory scrutiny to, for example, deep seek. How do you actually solve, I guess, the black box problem, this idea that, you know, you have these models, algorithms, whatever, and no one really has very good insight into what they're actually doing. And even if you have rules against something like redlining, the models can use proxy data to determine, you know, someone's race or someone's gender, income, whatever. How do you do that? Well, on those kinds of questions around discrimination, you're right, it's very challenging

to know what the intent of a model is. And, you know, everyone, there's a whole subset of people who that are really against when you anthropomorphize any model, but I'm just going to keep saying intent and desire. And we can have that conversation separately. So it's tough to know a model's intent, but you can know its impact. So to be clear, the Rays Act doesn't deal at all with those questions of discrimination. There's other bills pending in the legislature that do. And I think it's an important one. But in that and in the safety aspects, you can do a variety of tests to see how does it behave in certain circumstances. And you've seen outside researchers, you've seen companies themselves set up their models in what appear to be compromising situations and see how it behaves. So there

was one test that was done where a model was told it was going to be shut down, but was given access to what it thought was a company's email server. Oh, I remember this, yeah. And they had planted fake emails that the engineer conducting the test was having an affair. And so, So the model, after being told it was shut down, sent a message to the engineer saying, I am going to send these emails to your wife if you shut me down. So I don't know the intent of the model. I don't know exactly what's going on, but I know that that is behavior you probably want to work out of the model before it's released. It's pretty creative at the very least.

Yeah, to some extent, it shows they're working. You know, you mentioned Deepseek. That's an open source piece of software. They're in China. The people at Deepseek are not going to concern themselves with some legislation. in the New York State Assembly, anyone with an internet connection can theoretically download that model and run it on any server. Intuitively, this feels like it hobbles the American companies who have to abide by American laws, et cetera, and advantages open source models that are just

like maybe not even have a company behind them in the future, et cetera. Why does this not have a negative disparate impact on our own companies? So deep seek open source the model, but they are still a company that wants to profit and they sell things on top of it. So deepseeks available in the app store. If they don't pay fines in the U.S., we can have an injunction to take them out of the app store. And they want businesses to use it. So there is still a real reason for them to comply with the U.S. But on the flip side of it, how it could hobble these companies, we're not asking them to do much more than they've already committed to publicly. They made White House voluntary commitments in 2023 and 2024. They've had international summits like the Seoul, summit where they put

forward basically plans to do exactly this. What the Rays Act does is lock in place a floor so that when they're rushing for the next quarterly reporting or their next fundraising round, they don't have an incentive to cut out on safety. And I would say that the labs themselves, the lobbyists for the labs, did an estimate for what it would take to comply with the Rays Act. And they have every incentive to expand that estimate, say, this is a will hobble us, you know, we're going to leave. And the estimate they came back with was that it would require Google or meta to have one extra full-time employee. Oh, wow. How would the Raise Act actually interact with this new executive order that Trump just put out for a single national rule for AI?

So Trump is threatening to withhold funding from states and to sue states that do any sort of regulation around AI. New York already does a bunch of it. So regardless of whether the governor signs the Raise Act or not, we would probably be in Trump's crosshairs. For example, the governor this year put forward regulations around chatbots, requiring them to disclose that they are an AI model at the start and every three hours of continuous conversation. That seems so basic. So basic. And requires them to be on the lookout and to alert for when there's language that might indicate potential self-harm and to refer people to resources, which also seems really basic. But that would violate this executive order.

And so we're sort of in for a penny in for a pound at this point. and I think we need to stand up for New York families. Joe, I get notifications if I watch too many episodes of a show on Netflix, right? You can imagine people constantly talking to chatbots needing reminders, too. I feel like I've never had a three-hour conversation with a chat. If I ever have a three-hour conversation with a chatbot, Tracy, let me. I'm going to keep an eye on you in the office. You know, talk to us a little bit more about the politics of AI within Albany right now.

Is this the kind of thing? Is there a bipartisan sort of? anxiety. Talk to us about the vibe. Yeah, I think it's similar to what you see nationally, which is there are some, especially on the right, that just think the only thing that matters is how fast AI moves and not who it hurts. There are some, especially on the left, that don't want this technology put back in the box, right? And then there's most people in the middle that say, hey, we need to balance the benefits of it and the potential safety of it. And the Rays Act is

squarely within that realm. It passed with co-sponsors. who were both Democrats and Republicans, the majority of Republicans voted for it in the Assembly, and every single Republican state senator voted for it. So, Tracy, at the start, you were mentioning that proponents are the ones who are balancing safety and innovation. By that measure, everyone voting for it is a proponent. Why do you think Trump is doing this?

I love, why do you think Trump questions? No, no, no. No, no, honestly. It's a basic question, but I'm very curious to hear your answer. I, for so many reasons, wish I understood that man. better, but also feel like it would not be good for my mental health if I did. It is tough to understand because it's so different from his policy everywhere else. Right. He is putting forward tariffs and having this nationalistic, protective sort of anti-trade agenda on food and on

diapers and on toys. But then when it comes to AI. Pensils now. Pensils. He says we don't need so many pencils. My wife and I just had our first kid who's four months. So you see where my head's at with diapers and toys as we come into the holiday. Pencils are still a few years off. And then when it comes to AI, he just wants to push America's agenda everywhere and say no regulation whatsoever. So it's confusing. What I can say is that a number of the people behind the Super PAC targeting me are the same people that were funding his campaign. You had Mark Andreessen put five and a half million into his campaign. You had Joe Lonsdale put a million into his campaign. You had Greg Brockman, the president of OpenAI, put at least two and a half million into tearing down the

White House to build the ballroom. And so you have a lot of people around Trump at fundraising events and at other events who are pushing this agenda and he seems to have empowered. This is Caroline Hyde. And I'm Ed Ludlow inviting you to join us for Bloomberg Tech, a daily podcast focusing exclusively on technology, innovation and the future of business. Every weekday, we bring you the top headlines from the world's biggest tech companies. From finance to defense, AI to entertainment and from startups to the magnificent seven. We highlight the latest stories of the people and companies pushing the tech sector to new frontiers and the politics that shape global tech markets.

We do this all every weekday, then bring you the most important conversations and analysis in our podcast. Search for Bloomberg Tech on YouTube, Apple, Spotify or anywhere else you listen. Join us every afternoon on your commute home and stay ahead of the tech news cycle. That's the Bloomberg Tech Podcast. I'm Caroline Hyde in New York. And I'm Ed Ludlow in San Francisco. Subscribe today wherever you get your podcasts. So one of the things about the politics of AI, and I mentioned in the beginning, it's very multifaceted.

Because there are people that are worried about safety, people are worried about ethics, people are worried about water, people who are worried about electricity, people worried about job displacement, etc. As a candidate for office, how do you think about some of these other AI concerns out there, simply the safety stuff. Oh, there's a lot that we need to do. The chatbots, I think, are in some sense, safety when you think of what's happening in kids, but very much talked about in a different sort of conversation. And even expanding that out to AI's effect on our kids broadly in our education system.

You can imagine a really positive world where every student has an individualized tutor at exactly their level and teaches them in the exact way that they want to learn to supplement what they're getting from teachers. But what we have right now is we haven't updated our pedagogy, and so a lot of people think a signing an essay still teaches critical thinking. We need to do a lot on the education system. We need to do a lot on the workforce, as you mentioned. We need to do a lot on the environment.

And we're missing a golden opportunity here because our grid is extremely old, and we need to pay for the upgrades. And government doesn't really know how to do that. We've been passing off the cost of rate payers. Now you have an unlimited set of private capital, it seems. that wants to invest to build data centers, why aren't we using that to actually upgrade our grid and to require renewable energy and have it there? Instead, we're building these data centers but saying rate payers have to pay for the interconnects. It makes no sense.

Does this issue just AI in general, does it actually resonate with voters at the moment? Because in some sense, you're trying to get ahead of things, right? Because in the future, I know we said AI is all over the news and we see it everywhere, but in the future it's going to be more embedded in our lives. So you're trying to get ahead of that. Do people care at the moment? It's already embedded in their lives. They're seeing it. And it's not just, you know, their kids in school. It's not just the entry level unemployment at 9%. You know, we just saw a teddy bear that was sent out with Chatsy, BT, embedded in it that taught a kid how to find matches. It's Teddy Ruxpin upgraded. Do you remember that? I was walking down St. Mark's Street the other day.

I have to say, unfortunately, I thought there's a pretty clever. Someone set up a thing where there was like an animatronic puppet that had a camera in it and that it was embedded with an LLM. It was doing insult comics of the people who were walking by. Oh, my God. So like someone would walk by and they were like, so you think you bought it with a big Uniclo back? I was like, so you think you bought enough of Uniclo today? I was like, oh, this is clever.

It was super clever. My kids loved it. I was like, oh, this was like, this is like a. really clever thing. Wait, what did it say to you? What did it say? I think I stayed out. I do. I watched the other people go out. I stood, I stood on this. I was like, oh, this is actually, I don't know, my kids loved it.

We'll leave that as an exercise to the listener to walk by and report back. Yeah, what it would have said. What does it say to you? Are you optimistic about AI? Like, are you pro-AI in the sense that you think that this will be important, productive, positive technology that is important to continue developing? It can be, but only if the American people have a voice in how it develops. It is the technology that has the widest bounds of what could potentially come from it. And the only thing that comes close is nuclear energy and nuclear fission, right? So if you put yourself in the mindset of someone in the 1930s, you had one set of people saying,

nuclear fusion is coming, we're about to have clean, unlimited energy and live in utopia. And you had another set of people that said, we're all going to be dead from nuclear bombs in 10 years, right? And the reality is we've ended up somewhere in the middle. AI, we're at that moment right now where you have people saying basically that wide of a potential outcome. And it's up to us in our policy to make sure that the worst case doesn't happen so that we can have as much of the best case as possible. Like what AI could do for medical research, for curing diseases is remarkable. My mom has multiple sclerosis. Autoimmune diseases are some of the hardest for modern medicine to deal with. I am incredibly optimistic of what could come from some of

this research, it's just that the same capabilities that will allow us to cure diseases could, in the wrong hands, allow someone to build a bioweapon. And we just need to be thoughtful on how we got. I guess I'm worried about bioweapon. I'm also worried about very mundane things. Just like, the new Gemini image generator is just stunning to me in terms of the degree of fidelity, like how easily you really can't tell anymore. Like a year ago, you could say how that looks like an AI image. We were basically past that. Other things like bots and stuff like that, things that replicate our voice. These aren't like ultra-science fiction things like a machine that's going to like manufacture

a bioweapon or something like that, but they're day-to-day pervasive phenomenon that's sort of like, they're going to make me trust people less. They're going to make it harder to communicate like this or what do we do about that? Can we nerd out about deepfakes? Because this is a solvable problem and one that that I think most people are missing the boat on. So it's always been presented to us as like, oh, you'll have to learn. and how to see what's wrong with an AI image. Like, that's never going to work.

They're going to get better and better. Maybe we're already past the point where a human could see it. But that's not how we've solved these problems in the past. If you go back to the 90s, people said, we'll never have internet banking because you don't know that the computer on the other end is actually the one that you're talking to. And then we moved from HTTP to HTTP. That was a solvable problem. That basically same technique works for images, video, and for audio.

So there is a free, open. open source metadata standard that industry is created called C2PA, content credential provenance authority, that you can attach to any standard file format that cryptographically proves whether that content was taken from a real device generated by AI and or how it was edited over time. The challenge is the creator has to attach it. And so you need to get to a place where that is the default option built into our lives. So you see an image and it doesn't have that cryptographic prove, you should be skeptical. We should get to that place.

That would be the idea where it's so pervasive that the expectation is I can test very quickly whether the creator has attached that. It'd be like going to your banking website and only loading HTTP, right? You would instantly be suspect. But you can still produce the images. And I guess I've been reading a book called The New Age of Sexism and it's about technology and discrimination against women. And it has some awful, awful stories of school girls who's

classmates, like, you know, put them in porn videos and things like that. And one of the problems is there's no rules saying that you can't actually do that. So even if you know that it's fake, it can still be harmful. What do you do about that? Well, there are laws in New York State that ban it and in other states have taken action. And that's yet another reason why this AI preemption is such a scary thought. Because states are leading the way right now in stopping some of these absolute worst uses. I'm very excited for Congress to, actually solve these problems for there to actually be federal standards. I'm running on the platform of creating these federal standards. But until they do, stopping states from taking action,

like against deep fake porn of children, I mean, that's the stakes of what we're talking about. Can we talk about Palantir for a second? Definitely. So it's a rough segue. But what did you do there? So I joined Palantir as a data scientist in 2014. And I left is one of the five overall leads of the government business. I spent the vast majority of my time in the federal civilian side of it. So I worked with the Census Bureau and the Bureau of Economic Analysis on updating how they calculate GDP. We actually made a change to the way they account for spending around moving holidays, holidays that are sometimes in Q1, sometimes in Q2, like Easter. The paper I'm very, very proud of, and it's a rounding error of a rounding error on how we calculate GDP, but it did

actually lead to a change. I led our work with the Department of Justice to go after the opioid epidemic and to solve some violent crimes. I worked with Veterans Affairs to better staff their hospitals to make sure that veterans get the care that they deserve and need. I worked with the CDC to better track epidemics. It was about allowing government to make better use of the data that they already had to serve the American people. And what does Palantir actually do? I feel like we ask this question a lot. Well, I think it's because it's a fundamentally unsexy thing and people like to dress it up. But it's data integration and analysis.

It's making it different data sources that you have access to, talk to each other, be updated constantly. Like, Palantir was founded around the time when data lakes and data clouds were like the big thing of what's talking about. Yeah, exactly. But it's just putting...

Actually, I'm asking, what is it data like? I don't know what that means. Putting all of your data in one place. Okay. People can access it, right? And that was supposed to revolutionize everyone's ability to do anything. But just putting data in one place doesn't actually make a change. Some of the things that Palantirers really put at the forefront, like an ontology, right,

a view from on high of what each piece of data is supposed to mean can actually lead to better analysis. So I'll give you an example from my work at the Department of Justice. And I actually, I have two software patents for this project, which was we were helping them analyze banks behavior leading up to the great recession and how they were packaging mortgages into securities, and each security would have a loan tape. Here's the 1,000 loans that are in it. And some of those loans would not be up to the standards that were required. And that would occasionally be flagged before an issuance, hey, this loan isn't up to your standards. They would pull the loan out, and then their next issuance put the loan back in, right? And so if you could find that pattern of

behavior, you could maybe prove that they had knowledge that that loan wasn't good and we're still putting it out there. The problem is that eDiscovery software, all the data was there, but it was just taught to think of an Excel document as a document that a lawyer would read. And so there was no way in the software to track those individual loans. But if you think of an individual loan as its own object, that should be something that can be searched and tracked across the database, then that analysis becomes very easy to do. So that was something that we enabled at Palantir was putting this on of what's the right level for each object. Oh, a loan is a meaningful instance. Let's make it so you can search and analyze an individual loan versus a document. How has your experience at

Palantir actually informed your approach to government? Because we've done episodes on why government software is so bad. And I think you're one of the few politicians out there who actually knows how to code, possibly the only one. No, there's going to be a couple others. You think? Two or three. I don't know. We'll find them. I am the first Democrat elected in New York State at any level with a degree in computer science. I do think there have been nationwide like four or five congress members that have that. So they're out there. But my time of Palantir informs what I do in government, I think in two main ways.

The first is the work isn't done when the bill is signed, right? It's about the actual implementation. Everything Palantir does is about implementing things that have already been passed. And there's huge challenges in getting it to work. But the second thing is that basis in data and actually tracking your results and seeing how you've done over time. So few politicians will say, hey, that bill I passed two years ago, here's how it's working. And even more rare is here's how it's not. I did my town hall three weeks ago.

I once a year get up in front of my entire constituency, spend two hours answering any questions people have. We don't screen it at all. But I let off actually with, hey, here's a bill I passed a year ago. And the data shows it's not working. And here's what I've learned about that and how we're going to change that. How exactly do you judge performance by the government? Because I like the idea of like tracking, I guess, alpha in the civil service in some way. But would it just be based on the bill execution or I guess response rates from the public? How would you do that? So I'll give you an example. We passed a bill to raise the statutory maximum fine on telemarketers. By far my most popular bill. But the question is that's the statutory maximum. Does that actually lead to more fine? to higher fines to more negotiated settlements, is raising that of pressure.

And so we just looked at the Secretary of State's data of how much were the fines before that bill was passed and afterwards. And we found there were four times as many fines actually there. So raising the statutory max brought the bad companies to the table to actually negotiate and hopefully is changing that behavior. The example I gave of one that didn't work was around mopeds and e-bikes in New York. So, you know, this is, I'm going to be very local to New York. I know there's a nationwide primary.

It'd be nice if there were some talk of actual relevant things to the 12th district. So we have these delivery vehicles that are whizzing all over and people are scared as they see them go the wrong way, et cetera. And there's a lot of discussion about what we can do to make people safe on the streets. One aspect is that mopeds were already required to be registered in New York State, but almost never were. So I passed a bill that would require mopeds to be registered at the point of sale. So think about how many mopeds you think exist in New York City. the number that were actually registered when the bill came into effect this January was about 1700. Wow.

I feel like, by the way, this would be like one of those like hedge fund interview questions. How many mopeds are in New York City? I want to see how you brain thinks. And I think if your answer is 1700, the hedge fund's not hiring you, right? Like that is not what you would estimate. My thesis was it was that people were walking out of the store without registering and being told to do that. We looked a month ago. The number that are registered is now 1,400.

So it's not that people weren't registering when they came to the store. It's that registration expires after a year and no one's re-registering. So that was an example of like I had a thesis. I passed the bill. I think the bill's still overall good. People should register as they leave the store. But it didn't actually solve the problem. We were out to solve.

And here's now what I'm going to do about it. And I think we need a lot more of that in government. Wait, what are you going to do about it? Well, we're thinking about how we can encourage the actual registration, the re-registration over time. So one of the aspects is requiring all of the delivery apps to actually check that the driver are registered when they sign up and to make sure they keep that up to date. Makes sense to me. You know, it feels like we're like in this sort of era of everyday, you know, I mentioned

telemarketers. Like, we're just like waiting through the muck of like low level scams everywhere. Every day I get text messages. I know they'll say like from an unknown number of being, Joe, are you coming to a barbecue? And I'm like, I don't know. I might have like said, I might have agreed to eat barbecue with someone. That sounds like something I could have done. but I'm like pretty sure it's fake.

There's that you go on like Amazon and there's AI generated books about books and every author. Just generally, it feels like we're in a culture and economy and era of like persistent low level grift across almost every dimension of our lives. And I'm sure there's many reasons for it. But when people talk about a low trust society and what's happening, how much of this do you think is just because like we're inundated with people trying to rob us every day and somewhat digitally or otherwise. I think that's a big part of it. And the only way to solve that problem is you need actual enforcement.

You need there to be consequences. You know, sorry, just to follow up on this. Like I guess the reason I'm asking is because, you know, it's like someone running for us. They moved to the 12th district and like, I'm going to call out Trump's corruption. And there are all these like Trump is a lot, all this like big stuff. And it's, you know, whatever. many of these issues are like, you know, legitimate, et cetera.

But almost nobody in elected office seems to be just like talking about like, who's going to do something about the text message scams? Who's going to be doing something about the AI books? All of these like big national issues. But that doesn't affect me on a day-to-day basis the way this sort of like persistent abuse of technology constantly does every minute of my life. I think that's why podcasts like this are great because you get into the details. No, the top level news doesn't cover a lot of that, right?

I have a colleague, John Rivera, who has a bill in the State Assembly on requiring disclosure for AI-generated books on Amazon. Like that specific example is a bill that exists in New York. The telemarketing bill I passed also applies to text messages. And so we tell people how to report those and maybe get fines coming from that. I think this is key to both parties in 2026 is talking about just governments affect and trying to make life a little bit better. I think my other most popular bill besides the telemarketing one was my click to cancel

bill this year to allow New Yorkers... I would love that. It's now you're right. Really? Really? New Yorkers need to be able to cancel a subscription the same way they signed up for it. I have a Condé NASS subscription that I've been trying to cancel for like a year and I cannot log into the site.

I can't find the original email where I like actually signed up for it. And that's I think $12 a month that I'm just wasting. I mean, government needs to get big things done. Absolutely. But we also need to get small things right. You just need to make life better. I'm June Grasso, inviting you to join me for the Bloomberg Law podcast. Every weekday, we help you make sense of the legal stories that shape the nation and the world.

Listen for complete analysis of the biggest court cases, the latest actions from Congress and regulators, and the legal moves driving the markets, from corporate law to constitutional law, and from state courts to the Supreme Court. At Bloomberg Law, we go beyond the day's headlines. We speak with top attorneys, judge. scholars and policy experts to break down what the rulings really mean. We do this every weekday, then bring you the best conversations in our daily podcast. Search for Bloomberg Law on YouTube, Apple, Spotify, or anywhere else you listen.

On the East Coast, listen as you start your day. And on the West Coast, catch up in the evening. That's the Bloomberg Law Podcast with me, June Grosso. Subscribe today wherever you get your podcasts. Apropos of nothing, this is a completely random seg, but one of the interesting things about Eric Adams was his interest in digital assets and cryptocurrency and tokenization and all of that. What did you think about those efforts? Do they actually matter to the industry and to New York more widely, or is it just sort of

a pet project? I have not seen the direct results of what he was working on. that doesn't mean they weren't there. It's just not where I have been focused. I actually worked on a different bill this year around crypto in the legislature where New York really took the lead in creating legal structures for many of these companies through the limited purpose trust and the bit license. But now with new action at the federal level, they have said, you know, there's going to be federal licensing for these trust companies for stable coins. And they will defer to states, but only if the states have detailed rules in legislation or regulation, in statute or regulation. And New York's done it

almost all by guidance. And so I really fought for a bill this year to standardize what DFS had done in random opinions, put it into statute so that people know what the rules of the road are, and so that New York can keep its regulatory structure. It passed the Senate. Unfortunately, we didn't get it done in the assembly. And now most of the companies in New York are just applying for the federal charter. And we as New Yorkers are going to lose our ability to engage. So I think regardless of your broad view on crypto, finding ways to set rules of the road that allow for innovation and allow for us to have a say and how it develops is something everyone should get behind. Yeah, I'm looking at a map of the New York 12th congressional district. I think one thing that really stands out is not some weird snakey

district. It's just a real big square in the middle of Manhattan. Like people who don't live here might not appreciate this. This is like, this is prime real estate you're running for it. I can argue, I mean, this is, it might be the highest GDP district in the world or the contrary, or pretty close to it, right? Pretty close to it, if not the one. Yes, there's more Fortune 500 companies headquartered there than I think at least 37 states. You're a few blocks away from me, by the way. I'm in these village.

It looks like it just cuts off a few blocks. We'll fix it in redistricting. It must be weird, like, you know, thinking about going back to the beginning with the tech industry and criticizing you. And it must be sort of weird that, like, you're a politician who could actually. like talking like gigaflops and stuff like that because there's a lot of like anti-AI people etc whatever or open standards for digital encryption it must be kind of interesting that okay here is someone who wants to like hold the industry to some legislative standards who is frankly like

knowledgeable informed yeah yeah knowledgeable yeah that must be sort of weird for them i think it is and i think it's why all of their spending against me so far is backfiring because people are like, oh, yeah, who is this guy who doesn't understand tech who's doing uninformed things? And it's like, wait, it's the guy with software patents who worked at Palantir, who has a master's in computer science. There is a disconnect there. I think, you know, I have support from a lot of the people that are like my age that work in the tech industry, the ones actually building it because they see the power of what they're building and think there should be some protections for it. It's really those at the top who are, I think, primarily focused on profits that don't want government getting involved at all.

I have a theoretical question, going back to the idea of hedge funds, asking random questions to see how you think. If you had, I don't know, a Bloomberg terminal equivalent of government data, what would you be most excited about looking up? Or what correlation would you be most excited about finding? That is a great question. I'm going to stall at first by pointing out that I actually do have a certification in the Bloomberg terminal for my first job. So I'm imagining this realistically. So you know that all we do on the terminal is just look at two lines that correlate. I've proved correlation here.

I think ways of running more, well, not maybe running more natural experiments, but looking at over time how investments the government pays off. So we only budget year by year. And so you have things like investing in early child care that, you have things like investing in early child care that, boost your immediate who can be employed, but also like will help the kids if you have universal pre-K, et cetera. But that payback in the government, if you're just thinking fiscally, is not going to come for 20, 30 years. And as we budget year to year, that is, we think of that as a cost. I think if we had more ways of tying the effects previously in budgets with what we're seeing now, it would change radically where we would invest. Have you played around with the AI

code generation apps much. Yeah, yeah. So I, when I came into office, one of the things I really wanted was a coder on staff. I was like, there's so many creative things we could do of doing prototypes of government software or playing with the data, but we don't pay legislative staff nearly enough. It's one of the positions I've been most radicalized in office. But I had this idea of all these projects I wanted to get done. And finally, this year, Hunter College gave me an intern. That was a CS major. And so I gave her this list of like 10 projects. And she got through three of them and they were great.

But I looked at the other seven. And I was like, these are things that I know how to do. But it's not like worth my time at the moment versus all the things I have to do. But has AI coding sped up how quickly I can do these that becomes worth it? And I mostly use chat GPT, which I know Claude and Gemini I've played with. But I was able to knock out three of those projects this year. because of how more advanced the coding has become. It's pretty cool.

Alex Boris, thank you so much for coming on odd loss. That's a lot of fun. Thanks for having you. It must be kind of weird. You know, I said it near the end, but an AI critic that knows something about tech because there are a lot of AI critics out there, and I don't get the impression.

A lot of them are particularly actually well-informed and do not actually understand the sort of strongest version of the argument that they're running against. So it's interesting that this pack, this pro-artificial intelligence pack has decided to target someone who I think a lot of people listen to it's like actually he knows what he's talking about he sounds pretty reasonable he's also the first one they're targeting right yeah so it'll be interesting to see who they go after

yeah after him the other thing I was thinking is if you think about one of the threats to big tech's business models it's it's always regulation right or at least they say it's regulation they seem to hate regulation or over regulation, I should say. So if you think about proposing some basic safety guardrails and privacy, that would seem like a good thing to me in the sense that maybe you could restore some of the trust or faith between the general population and the big tech companies and then you wouldn't get these massive fights about everything else. But that's probably a long shot. It does seem like I have to say that while a lot of what he's, He said and made a lot of sense.

I sort of sounded very reasonable. You know, I do think that it's legitimate concern. Like, these issues aren't going to hobble Deepseek. Now, does Deepseek want to be in the app store? And can U.S. laws enjoin them from being in the app store? Sure. But they're not going to hobble Deepseek's development of a model. And if they want to be a dominant player in every country but the U.S.,

and they're completely unconstrained and the American advantage, labs are constrained. I do think that is legitimate concern that the industry might have. I also think it's a probably legitimate concern that you don't want to disincentivize the reporting of safety issues, that if a company gets penalized for acknowledging that something in their model like violated some safety line, you do not want to have the risk of like, oh, we're just going to ignore this. And I think it's reasonable that the industry might be concerned that regulatory, could benefit the incumbents. You know, he argued against that.

He's like, no, like this is not going to affect like startups, et cetera. But there is always that risk that a monetary fine is something that's very easy for big companies to pay and more marginal ones are more difficult. So, you know, these are very tricky things. But as, you know, on the service, more than on the service, an extremely knowledgeable guy who, what he says, none of it sounds particularly ridiculous. And I have to say, if I were running an. primary that had like 25 people, including like a Kennedy Air and some guy on Twitter

who like posts Orange Man Bad a lot of times. That sounds like a good way to stand out from the crowd to be the AI industry's number one target. You know, I agree with you on the incumbent point, but just in terms of U.S. versus China, I think people forget that China has its own restrictions, quite severe ones on censorship, right? And trawling, I guess, every single communication in the world in case someone mentions a certain cartoon character. Seems like that's a big regulatory burden as well. So I don't know. I find the competition aspect of it overhyped in some sense.

But anyway, we could talk about this for ages. And a lot of these issues are clearly like ongoing and no one has the, the solution. Yeah, we're trying to work that out. All right. Shall we leave it there? Let's leave it there. This has been another episode of the Odd Thoughts podcast. I'm Tracy Alloway. You can follow me at Tracy Alloway.

And I'm Jill Wisenthall. You can follow me at the stalwart. Follow our guest, Alex Boris. He's at Alex Boris. Follow our producers, Carmen Rodriguez at Kermann, Dashel Bennett at Dashbot and Kail Brooks at Kail Brooks. And for more Oddlots content, you can find all of our episodes in the daily newsletter at Bloomberg.com slash oddlods. And you can chat with fellow listeners in our Discord 247, Discord.g.g. slash oddlots. And if you enjoy Oblots, if you like it when we do these AI episodes, then please leave a

positive review on your favorite podcast platform. And remember, if you are a Bloomberg subscriber, you can listen to all of our episodes, absolutely ad free. All you need to do is find the Bloomberg channel on Apple Podcasts and follow the instructions there. Thanks for listening. I'm Francine Lacqua,

an award-winning journalist, and I've got a new podcast, leaders with Francine Lacqua from Bloomberg Podcasts. I've interviewed everyone, from Heads of State to fashion icons about the news of the moment. But I've always been curious who are these people as leaders. I don't think there's one right way to be a leader. Make decisions.

A poor decision is always better than no decision. Listen to new episodes every other Monday. Follow leaders with Francine Lacroix wherever you get your podcasts.