Central Air - We All Live in Florida Now

Episode Date: April 1, 2026

On this week's show: Marc Caputo, White House Correspondent for Axios, joins the podcast to help us understand how Trump and his advisers are deciding what to do with the Iran war, and how they are preparing (or not) for the domestic political blowback from an extended disruption in oil markets. We also get his view from south Florida on the ongoing Republican dominance of that state — what the party did right to win solid majorities of Florida voters, and whether they face any danger from Trump’s national unpopularity and a cost of living crisis that, in Florida, takes the particular form of high housing prices and skyrocketing homeowner’s insurance costs.Plus, Megan came in for a two-minutes hate this week for describing how she uses AI in her writing process — primarily, in the way one would use a human research assistant. We talk about the right way to use AI as a journalist, and the roots of the anti-AI fervor among journalists.Sign up for updates from Central Air at www.centralairpodcast.com This is a public episode. If you'd like to discuss this with other subscribers or get access to bonus episodes, visit www.centralairpodcast.com/subscribe

Welcome to Central Air, the show where the temperature is always just right. I'm Josh Barrow. I'm here, as always, with Megan McArdle, columnist for the Washington Post, and Ben Dreyfus, who writes the Substack newsletter, Calm Down. Megan, I have some fun personal news. I'm going to see the opera tonight. Oh, Josh, I'm so excited for you. My friend Phil Chan heard me say on the podcast that I'd never been to the opera. Phil is an accomplished opera director and also ballet choreographer.

And so he very kindly decided to rectify that. We're going to see La Traviata at the Met here in New York. Good choice. Good venue. Have you seen that production? I have not seen that production, but I like the Met and I like La Traviata. Yeah. I assume that it's good there.

Usually it is. It's also not as long as I thought. The runtime is like two and a half hours. So it's actually shorter than a lot of Hollywood movies these days. It's not the ring cycle. But so I'm going to go see that. And then actually, we're going to do a substack live video chat on Thursday with Phil. And so we're going to talk about the opera that I saw.

And we're going to talk about how opera can be saved from the likes of Timothy Shalameh. And, you know, more broadly, Phil has actually had some interesting things to write about how, you know, you can get offended at what actors say about the opera. But ultimately, opera to survive needs to make itself relevant to new generations. And he has some ideas about how to do that, including a, you know, a very well-regarded and also. So a transformative production of Madama Butterfly that he did for the Boston Lyric Opera a couple of years ago. Well, I promise to brush up by listening to Traviata before we do this. Ben, I forgot. Have you seen an opera? Of course not. I don't care for the Italians as a people or in their language.

Don't need to hear it. Wagner exists, Ben. Yeah. The Germans are even worse. Don't get me started on them. There are French operas. Yeah. I'll, one day I'll go to, I'll go to an American opera in English, the good old, the king's language. They exist.

Okay. Well, we'll see, we'll see if Phil can, can sell you on Thursday. What's the opera he goes to in Pretty Woman, where he does the thing with the necklace? What they see in Pretty Woman is La Traviata. So I'm going to have my own Julia Roberts experience. Oh, my God. That's, that's, now I'm way more excited. Are you wearing the red dress?

Because that dress is like every woman who saw that movie wanted that dress. And I think that you should honor our clothes list. It's at the dry cleaners, unfortunately. Rent the runway, Josh. But so that will be Thursday at 1. And if you happen to be listening to this podcast after Thursday at 1, we're sorry we missed you. But there will be a playback available to paying subscribers so you can see that conversation at your leisure. But this week, let's start with Iran.

a significantly heavier topic than the opera. And for a conversation about the war and the Trump administration's decisions about the war and how we got here, I'm really excited that we have Mark Caputo with us this week. Mark Caputo is White House correspondent for Axios and has been doing a lot of Axios's reporting on internal decision making within the Trump administration about this war. So Mark, thank you for joining us. Thanks. I would like to talk about the Brindisi, which is the wine drinking song from La Traviata, which is fantastic. Okay. And as an Italian, I'm offended by Ben Dreyfus' hatred for my people.

When I made that joke, I saw your face and I went, oh, Jesus. Yeah, I'm like, what the fuck, man? Lots of fool. Anyway, lots of names ending in vowels on the show this week. But so, Mark, first of all, you know, for listeners that don't know, you're based in Miami. Yes, sir. And the funny thing about this administration is that South Florida is quite a good place to report on what they're up to and what the internal intrigue of this administration is. So I guess that's convenient for you.

It is to a degree. I mean, albeit Palm Beach is a little different from Miami. You know, I think we're, I think Palm Beach is farther from Miami than Baltimore is from Washington, D.C., which I frequently remind my bosses of. But, yes, this is a Florida in exile or Florida in practice White House, considering all the different ranks and how they're populated there. This week, I've been just sort of slagery. black-jawed looking at the latest news out of the administration where they're basically saying, well, you know, if we don't reopen the Strait of Hormuz, then so be it, you know, maybe the Europeans can take care of that. And it's, it strikes me as among other things a really bizarre domestic

political calculation on the part of the president because gasoline's already gone up a dollar a gallon. We may be looking at $5 a gallon gasoline in June July, just saying, well, gee, that's the British's fault, is not, I think, going to cut it for him at home, let alone abroad. But do you have a sense of, you know, day-to-day, how they're even conceiving what this war is for, what their objectives are, and what they consider to be satisfactory outcomes? Yes. I mean, first thing, sort of a category error is to think that Trump is making a calculation domestically at this point. He's focused on the war in sort of how to prosecute it and how to get out. I mean, the administration has been very clear, of course, until Rubio yesterday gave a slightly different answer for the four

objectives that they're trying to attain militarily. And the struggles to the degree they acknowledge them that the White House has is rectifying the apparent. And I'm going to say apparent because there's some reporting suggesting it's not tactical success, military objective success on the battlefield. And squaring that with the strategic failure. of embarking on the war in the first place. And how the White House sort of muddles through that is what has all of us sort of, as you said, sort of slack jawed at how they're approaching it. So in terms of tactical success, the idea is, you know, we've destroyed a lot of their, you know, missile launching capability. We have presumably set back some of their nuclear operations

for some period. But even just looking narrowly at that, and setting aside the impact on global oil markets. I've been seeing this news about this idea that we might send commandos in to try to seize the uranium. Yep. And this sounds like something out of a movie that you have to have them on the ground for several days to get these giant things out. You have to build an impromptu airfield in order to lift them out of Iran. Sounds like it may well not work if they try it. It also sounds like they may not try it because of the logistical difficulties there.

Building airfields in Iran didn't go well for us that one time. Which time are you referring to? Ninety-79. The failed hostage rescue, that's what happened was there was a disaster at the impromptu airfield they built. So you can define tactical success as destroying certain weapons systems and that sort of thing. But doesn't that just put us back in the situation we were in before where in a year, then we're worried that they can build a nuke again? And then there has to be another set of military actions. Are they thinking that far ahead?

Yes and no. They are thinking ahead to the degree that some in the administration are adopting sort of a, an Israeli concept, it's more common that you'll hear IDF types talk about it and neocon military types here in the United States side. The phrase is mow the grass, which is you go there, you bomb the hell out of things, you set them back, then they come, in this case it would be the Iranians, they would come back, they would start to rebuild, then you go back in and bomb them again. And that conversation is occurring, albeit not at a fever pitch level. But that's among the many options.

The thing with Trump that's so sort of complicated here is there was a dash in a war. They made the conscious decision to launch a surprise attack. And as a result, didn't sort of lay the intellectual or informational battlefield with the explanations for why we're going in, why we need to do this, well, how things will shake out, what will happen to gas prices. And as a result, everything sort of looks haphazard and surprise. If you take the Pentagon briefings as being accurate, the actual process of limiting the Navy, the missile capacity, the firing capacity, the industrial capacity of Iran is proceeding a pace and is setting them back a bit. But as you're seeing now, Pakistan is talking about there in communications with China to come up with some sort of solution. Most people across the globe think that in order for there to be sort of some semblance of a lasting peace, you're going to need to know. an actual negotiation, an actual sitting down in a meeting of the minds. And that hasn't happened yet.

And so when you say it's a category error to look at this in terms of the effect on domestic politics. Or to think about Donald Trump as making a calculation about thinking about the sort of second order effects here. Well, but he's clearly making calculations about the effects of the markets on domestic politics in other ways. The statements and the timing of the statements he makes look calculated to push the stock market up in an effort to hold down the barrel price of oil. He is. With diminishing success. And, you know, as we tape this on Tuesday, March 31st, West Texas intermediate is over 100 a barrel. So he seems very cognizant to the fact that negative effect on stock prices and positive effects on oil prices are a significant political problem for him.

It sounds like if, you know, if it's that Israeli approach that you're describing where we're basically going to be in a permanent state of, of common. conflict with Iran, you know, not always bombing, but always in this hostile position where they're expecting more bombing. It seems like they should be expecting to go into the midterms and beyond with $5 gasoline, which doesn't seem like it can be politically sustainable. And I'm, you know, people talk about we're a net oil exporter now. And so obviously this doesn't have the same degree of negative effects on us that it has on a country like Japan. But, you know, just because some oil producers are going to make a windfall in Texas doesn't mean that U.S. consumers will be okay. with high-priced gasoline and also the negative global effects have knock-on effects in our economy. I mean, this is going to cause them a world to hurt. It is. And if you would listen to Scott Besson, the Treasury Secretary last week, he had said at the cabinet meeting that Democrats and critics of the administration are underestimating the American public's ability to experience short-term pain in return for long-term gain. and I'm wondering if it's the administration that's overestimating the American public's interest in that. You're asking if they're prepared to go in with $5 gasoline.

They're not admitting it when you talk to them. They claim that there can be a resolution and things can come down. Everyone I talk to seems to know about how oil and oil markets work and the price of gasoline don't think it's going to really reduce that much. I can't say it's going to be $5 gasoline. I'm not saying it's not going to be. but they're engaged in sort of very short-term thinking about this. And right now, the short-term thinking is, what is the best way to withdraw from this conflict within the four to six-week period that Trump had initially and generally laid out? And we're what in, I guess as of today, you say, what, we're March 31st. I guess April 11th brings us to the six-week period. That's still the sort of target date for stopping the operation. and we might want to start thinking of it in terms of this phase of operations. I don't want to say that the mow the grass scenario is a guaranteed long-term strategy.

The complicated thing about Trump is he is throwing out every possibility. And this is what he does. And his staff has mythologized him and his advisors as this sort of grand strategist who's intentionally doing this and has some grand plan. And of course, a lot of other people as critics especially think it's just sort of madness and improvisation. Megan, what do you make of this? I mean, you know, as we get into April, the shortages that are already popping up in East Asia will start affecting the U.S. market. Yeah, I mean, the best argument for this is that you can degrade Iran to the point where they can't really restart their nuclear program because they can't afford it. And that's the argument that you hit their dual-use stuff like steel plants.

I'm not sure I buy it, and it's definitely not going to happen on a timetable that will be domestically politically acceptable. The news Monday night seems going to, you know, still planning to end on the timeline of four to six weeks without reopening the state of foremost and then he was going to ask Europe to do it and then Gulf states. I honestly don't understand it. And I feel like maybe, Mark, can you sort of even explain to me what the reasoning it here is? It seems like, I don't know, maybe you should give it a couple more weeks to open it up. Seems like that should be a priority opening it back up instead of asking the Greek Navy to do it. Yeah, I've also been surprised by the idea of four to six weeks is actually, because like the president is famous for saying, you know, in two weeks and two weeks, I didn't realize that four to six was a real number that actually related to the calendar and the clock. I mean, it's generally been what he's generally said.

And to your point is things with Donald Trump are set in fast deadlines until they're not. And just Trump has essentially said everything to advisors. Reporters like me and the reporters of the Wall Street Journal who did that report on Monday night about how Trump has talked to advisors about maybe I should leave the straight of Ormoose. Well, yeah, he said that. And he's also said to advisors, maybe we should have a land operation with special forces to get uranium out. And he's also said to advisors, maybe we should take Carg Island. The oil depot are 90% of the crude in Iran flows. And he's also discussed the idea of taking La Rock Island and Abu Musa and the Tunvi Islands that would help control the state of Hormuz. And so he's just saying lots of different things. And there's a sort of a media story here is that right now we don't see a lot of. new news, right? We're bombing stuff, but that's not really new news. And the Defense Department is only sharing so much information anyway. And so we're left with, okay, well, what's the, what's the president talking about? And what the president's talking about is he's talking about a lot of

things. So this is sort of a roar shock test, you know, different people hear different things from a Roshaman effect, which is, you know, different people hear different things depending on their point of view. And it just makes for an incredibly challenging information environment to operate in. But this is sort of trumping Trump is just kind of saying what's in his head, partly because he doesn't have a lot of impulse control and maybe because he's also strategically ceding the information space with all of this noise so he can get to a place where he's ready to really make a decision. One of the things that we have to understand about what we've seen from Trump so far when we judge his actions more than his words or his actions in tandem with

his words is this. When Trump talks about making a deal on one hand, and then he starts massing forces with another, he generally winds up executing more military action when he doesn't get the deal he wants. And we are massing more forces there. So non-zero chance, we're going to see a different escalation, or as I mentioned before, maybe this is Bays 1, a different phase of this conflict. I mean, one infuriating thing about this that is now just treated as part of the wallpaper is that we're all sitting here waiting for him to make these decisions, which he does in this opaque and, you know, scattershot manner. In theory, you know, it's not supposed to be just his decision here. You know, the Constitution says Congress declares war. I realize we haven't done that since World War II, but Congress has to appropriate funds. They have to authorize the use of military force, et cetera. And we've seen some complaining from even Republican members of. Congress that they feel the briefings they're receiving are insufficient. But so far, it doesn't really seem to be any kind of limiting factor on the administration's operations. They have asked for this

supplemental spending bill to finance this. How long can they go on without getting a supplemental spending bill? That I don't know. I'm I'm remiss in not covering the Pentagon's budget. I mean, these are the guys famous for the $100 hammers. So presumably there's, money squirled around somewhere. And Donald Trump is not shy about declaring emergencies and moving money around the budget. One of the things to remember about Trump, what makes this kind of phase of the American Republic different from the others is you have an executive who's been impeached not once but twice and has been there and done that. And he, accepts certain limitations.

He had said to the New York Times famously, the ultimate limitation on me is me. And he's willing to do what he wants to do and basically tell Congress, well, if you don't like what I'm doing and what I'm doing is right in my opinion, go ahead and impeach me and throw me out of office. And he knows they have to get to 67. Now, does Donald Trump want to get impeached?

No. Does he want to have a trial in front of the Senate? No. But is he keenly aware of the effort it would take to rein him in when he really wants to do something? Yeah. And he doesn't have a lot of respect for a lot of the House members. Like Nancy Mays starts complaining from South Carolina about the military action. Well, yeah, that's after she doesn't get endorsed.

So he doesn't really kind of factor those things in. Well, although if he wants that supplemental spending, Billy, can only lose one Republican vote. So he needs Nancy Mace. He's not going to get Thomas Massey. He does. And we're going to see in a few months whether that comes to pass. But I honestly do not know that one. I'm going to have to check out because they still can't really open DHS fully.

Right. That section of the budget is still kind of closed. And he just took executive action there to pay TSA agents. And again, there's that limiting factor of Congress. If Congress wants to stop them, they can step up and do something and they're not. What do you make of the dissension that's coming from certain areas? of MAGA media complaining about this. Megan Kelly, for example, is extremely exercised. You know, you have to stop this war immediately. It's going to destroy your administration, et cetera. And there

are certain prominent media figures who have been critical, often the same prominent conservative media figures who have been critical of Israel. I don't see signs yet that that's flowing through into the polling data. Is it flowing through into MAGA world in terms of, you know, people with, you know, who are ordinarily considered, you know, influential inside the tent having those concerns? Well, I mean, the people who are opposed to Tucker Carlson's of the world just are. And the effect it's having is that Donald Trump as a result is talking more of Mark Levin, you know, right? So it's just kind of how he is. The, I do think that the Tucker Carlson's and the Megan Kelly's and the Tim Pools and the like of the world, the Steve Bannons, are outside of being representative of their sort of more intellectual, social class, are looking at. the polling data of younger voters, the promises to the degree they feel that they were made of Trump and especially J.D. Vance and looking at the future and just thinking, this is not a sustainable course. And Trump's sort of true to form doesn't really give a shit what they say.

He's still looking at the polling. The polling still shows 85% or so 90% of Republican voters, if not MAGA voters, are still with him. One of the things I've urged people to remember is the Republican Party is the party of the U.S. military since the Civil War. And there is something latent in the DNA of the GOP where when the commander in chief, the Republican commander in chief, says it's time to go to war and bullets start flying and bombs start dropping, their pulse starts to quicken and they kind of fall in line. Now, what can change that? What Megan Kelly and company and what all of us sort of see who have to drive, which is every day we drive by the scoreboard of the Iran War, and it's showing higher and higher prices at the pump. And Trump voters drive trucks, and they drive

internal combustion engine vehicles. They're not Tesla drivers, by and large. And as a result, there could be a bill that has come to do, come do politically speaking, but they just haven't reached it yet. That pain threshold I talked about with Besson earlier, they haven't hit it yet, but I do share the opinion that that is coming. And you've even seen up until now with all the special election results, well, yeah, sure, maybe Republicans are behind Trump 80%, 85%, 90%. They're not showing up to vote. And it's a political problem heading into this midterm. And Trump kind of doesn't care as much about that as they do. Let's take a quick break. And then we'll take a quick break and then we're going to come back with Mark Caputo to talk about Florida politics. This is Central Air.

So, Mark, I've been interested the last few days watching Rhonda Santis's Twitter presence, the governor of Florida, who we used to talk about a lot more than we do lately. Doing this sort of, you know, in addition to talking a lot about sports, and he's, you know, he's going to be done being governor in a few months. He's been doing this sort of dispassionate market commentary, including there was one post that was just the 10-year treasury yield. And he's like, gee, that's getting awfully high. Mortgage rates are getting. high too. What do you, I mean, obviously he's, you know, he was never happy that Trump got the nomination instead of him and he ended up on the outs from Trump World. What do you make of that sort of needling from Ron DeSantis? I mean, DeSantis is sort of playing both sides. At the same time, he was saying that. He signed the legislation to rename Palm Beach International Airport after

Donald Trump. He also signed the legislation the year before cutting the city of Miami and the County of Miami-Dade out of any local government oversight, control, or say in the massive new plan for the Trump Library Tower in downtown Miami. So he does understand still who the boss is. At the same time, Ron DeZanis obviously wants to be the boss, right? He's looking at what's happening. He, sure, he got sort of beaten and beaten handily in the GOP primary. and it would not surprise me to see DeSantis give it another run in 2028. In part, you look at J.D. Vance, like, how strong is J.D. Vance going to be? And guys like Ron DeSantis look at that, he's been to the dance before.

He understands a little more of what it takes. DeSantis looks at Vance and might be sizing him up as someone that he could sort of beat as a sort of a MAGA insider and a MAGA outsider at the same time. basically the promise of the original DeSantis campaign that he never fulfilled. I'm kind of skeptical of that because a major reason DeSantis' campaign fell apart was because of DeSantis. But the top reason was because he ran against Donald Trump and just didn't understand like no man, modern politics are cults of personality and there's no cult of personality like Maga and Maga controls a GOP. We'll see what happens in 2008.

You're still going to be very short in 2028. Same here. He's taller than I am, which is going to, you know, that's one of the reasons I'm never going to run with president. Megan, it's sort of been interesting for me watching this. I mean, because, you know, again, I talk about $5 gasoline. What happened in 2008 was that, you know, as the popularity of the Bush administration was falling apart, there was value in getting a certain amount of distance from him messaging-wise if you were going to be a Republican running for president. This party has so thoroughly purged anyone who has real distance from Donald Trump. And people love to talk about Trump's floor in terms of the polling. But his first term, he presided over a strong economy until COVID hit, which I think the electric correctly interpreted as a deus ex-Machina, not his fault. And so we haven't seen what it's like for Trump to preside over a recession or preside over an energy crisis that can be very squarely blamed on him. And so I don't know that DeSantis is the guy to do it, but I'm, you know, I'm wondering about the space that might open up in a Republican primary for someone who can

credibly say that stuff was not all my fault. I also wonder about that because I think that I think this was going to happen anyway. You know, when he leaves, he's going to be in his 80s. He's obviously not going to be the eminoles gris of the party for the next 20 years. And I think he's going to leave more unpopular than he did, certainly by his first re-election campaign, whether he'll be more unpopular than he was after January 6th. Different question. So I think yes. I think the idea that this was just Trump's party and was always going to be and that the race was to be his error was wrong. And people are starting to figure that out. I mean, in 20 years he'll be dead. But I had been, you know, sort of imagining the next several years after his presidency that he would still be this presence on social media demanding that Republicans do things or not do things endorsing candidates.

and the people in the party base would deeply care about that. You know, if he becomes unpopular enough, that can change. Yeah. And four years in the wilderness does a lot to make people rethink their previous choices. Yeah. I mean, you bring up a fascinating thing that I hadn't really thought about. But when he left the first time, a lot of people, he was on truth social and all of the people dutifully went to truth social. And once he's no longer the head of the party and he's just an unpopular person posting on truth social, I wonder how long they'll, and they'll stay there and then I'm coming back to X the Everything app.

Oh, I think that platform's going to collapse pretty quickly. Right. But then he's then he's just talking to Tim Poole or whatever on the truth social. Mark, I'm wondering, you know, to go back to DeSantis, what's on tap in Florida right now? In particular, there's been this expectation that there will be a congressional redistricting similar to what Texas did. And my understanding is they've been waiting for this pending Supreme Court decision that's likely to strike down part of the voting. Rights Act that governs how you have to draw congressional districts. But are they, you know, are they really going to try to do an aggressive remap there? Because I'm wondering if it's a bad

year for Republicans and you try to draw a map to pick up five additional seats or something, whether you end up spreading the Republicans in Florida too thin. I know, I know it's hard to imagine Democrats winning in Florida based on the last 16 years or so of performance, but it has happened. It has happened. It's hard for me to see, not impossible, because in 2010, Florida voters approved by 62%, and it takes 60% to approve constitutional amendments in Florida, amendments of the state constitution that put very clear limits on gerrymandering. And in very clear language, it says the Florida legislature cannot draw legislative in congressional districts with the intent to favor or disfavor an incumbent or a political party.

There are exceptions it makes for race in some respects. And if the Florida Supreme Court, which is almost entirely now, Desantis appointed, decides to ignore the plain language of the state constitution, I could see this passing more easily. Well, isn't that what they did with the last map? I thought the last map was also in violation of the constitutional amendment by any reasonable read, but the court just ignored it. No, the last map was none. They're in one clear case, not to get too much into the weeds, in Tampa Bay, I think they bent the understanding of favor, disfavor, political parties in incumbents. And also they bent the understanding of what was called compactness. That is, they allowed a district to cross a bridge and stretch between St. Pete and Tampa, the Tampa Bay. But that aside, if you look at the maps of Florida, they're pretty clean districts overall, especially compared to some abomination, relatively speaking, like Illinois, right,

where they're just clearly, clearly racial, or better said, partisanly gerrymandered. So if the Supreme Court looks at the Florida Supreme Court, looks at the plain language, the state constitution, and if they are not limited by clear sort of direction from the U.S. Supreme Court, I don't see how there can be a massive change in the congressional districts just right off the bat. Secondly, in a state as big as Florida, where we have a tendency to staycation a lot more than perhaps other states, and where a lot of us drive, the increase in gas prices, the related increase in inflation, the increases we've seen statewide in insurance rates, the decrease we've seen in quality of living, make this a far more challenging environment for Republicans to operate in.

And if you were able to wave a magic wand, and we'll put aside the anti-Gerrymandering amendment, and you were a Republican, there are those who are saying, maybe we don't want to fuck around with these districts too much, because instead of creating more Republican districts, we might be endangering more Republican districts by making these red districts a little more purple. And this could be more of a, I don't want to say a blue wave year, a more of a purple wave year. And it might be counterproductive. Because generally, if you are playing around with congressional districts, the,

the places that you can really start to chip away at, and the one that's going to be easiest to do is Sheila Trefellis McCormick, who got indicted, right, and is in Broward County. Her district is a little sort of what you would call a racial gerrymander. It sort of looks like a crab and it's got two claws. If you cut off those claws, though, there are a lot of black voters and black voters tend to be about 90-10 Democrat to Republican. do you want to send a 90% Democrat voting public into another district? If you do that, you don't make that other district more safely Republican. You make a little less Republican. So there's that kind of

calculus, that algebra that takes place that could make this more complicated as well. So number one, state constitution makes intentional political gerrymandering far more complicated, essentially bans it. Number two, even if you didn't have that, in this environment, if you started changing those districts, you might make things counterintuitively or not intentionally more beneficial for Democrats, not less beneficial for them. The other big political news I've been seeing out of Florida is that the state may get rid of non-school property taxes for permanent residence? It might.

It takes 60% to approve of something. It's a haphazard idea. If it makes the ballot, that's really difficult to do. Generally speaking, when you run an election, especially in amendment election, if you're spotting your opponent 10 percentage points. That is all they need to do is get 41% the losing side and they win, right? And that's just very challenging. School boards, local cities and counties are going to come out and force against that. I don't know how well that's going to fare. Okay. I did think that it was interesting that you, Mark, like what you were talking about with, you know, whether it's a purple year or a purple wave, you know, if Democrats do a little better in Florida specifically. And I guess it ties into, I'm not entirely sort of like certain what the GOP is particularly in places like Florida.

What the campaign argument is going to be in this fall. Like, it seems like a lot of things are piling up for them that aren't going terribly great. And I guess I see them all talking the same sort of talking about how Democrats are lunatics. But is that going to be it? Like, is that, have you seen, have you heard anything going on that is different than that? No, talking to, I don't want to. to get ahead of one of my colleagues at Axios is going to be writing about this very thing. What we're trying to get a handle on is what is the GOP message? Because just when you put things

on the spreadsheet, there's just not a lot of wins and deliverables. Last year in December, I wrote a story about how the Trump White House was thinking like come April 15th or so, right, which is in essentially two weeks, the economy is going to start to hum in the United States. The benefit of the tax cuts and tax package are going to come in. People are going to have more money in their pockets, interest rates are going to be coming down, you know, everyone's going to be living off the fat of the land. Happy times are here again. The war in Iran has totally changed that calculus. My sense, you know, because we went through several years of Democrats in the North looking at Ron DeSantis and looking at Florida and sort of going, how the fuck is this guy winning so much? How is he getting reelected by almost 20 percentage points? And, you know, partly, you know, the, we went through a period where Republicans were doing, were greatly improving.

they're standing with Hispanic voters generally, and that mattered a lot in Florida. But my other sense of the piece of the picture that wasn't obvious when you're looking down from New York was that people were pretty happy with the provision of government services in Florida, that there was a sense that the state government was doing a good job of the blocking and tackling on stuff. DeSantis passed several bills that raised teacher pay and, you know, had initiatives on Everglades restoration that were popular and that sort of thing. Is that still an advantage for the GOP? or is it now like, because now what I read about is people are really upset about homeowners insurance premiums and that it's not just the national affordability problems, but that Florida has additional problems of its own stacked on top of that.

Is that now a relative point of weakness for Republicans, or do they still have an advantage as the party of a government that people think works better than the government in states like California and New York? I think a little bit of both. DeSantis is more popular now in Florida than Donald Trump is. and the state government is viewed as being more successful than the federal government. And so if you make this an election about Ron DeSantis' legacy, more beneficial to the Republicans, you make it about Donald Trump's legacy, less beneficial. The thing about DeSantis that people just have to remember was, and I really, really got blackpilled by the social media commentary that crest that started a really crest in 2020 over how COVID was covered, how it was discussed on social

media, and specifically the social media panopticon that occurs with the sort of the hive mind that reinforces its perceptions and its institutional demands on everyone else. And so reporters like me when we were covering COVID and thinking, wow, things look really shitty in New York. And all of the national media coverage based in New York was basically communicating people. This is the plague. Wrap yourself in cellophane. Hide under your bed. Shut everything down.

Don't go to church. Don't breathe. And in Florida, DeSantis, and I remember being at this press conference in like March of 2020, March, April. you know, early March or mid-March, early April. And he was saying, you know, I'm reading all these things about COVID. And this thing seems to be a lot more like the flu. And it's going to be endemic, he was saying. And he'd listed all of these things that wound up being true.

But at the time, it conflicted with the expert class and the hive mind on social media. And anyone who even reporters who even wrote about that were just attacked as like pro-murder. And over time, DeSantis's stubbornness, his attention and detail, and his refusal to be cowed, won't up being the right bet. Now, some of that's because of his intellectual acumen. Some of it may be because he's lucky. I'll let people to decide that. And as a result, Florida boomed because everyone came down like, yeah, I'm going to Florida. Like, New York sucks.

And so we saw this real influx of people. And you saw the narrative change over time. And so that was really DeSantis's strength. The problem that he wound up having is that he was just sort of like after 2020, 2020, 2021, 2022, where he sort of rode that coattail. He was sort of the jacu's returning to high school and want to relive those glory days of being right over COVID. And, you know, by 2024, everyone had kind of moved on. And that's the sort of dynamic that he's in now.

But that really sort of helps explain why he seemed to be doing. doing everything right and why it was so popular because he was sort of laissez-faire. Like, there was a fatalistic aspect to it, which is, look, you know, you might die, COVID. Sorry. You know, trying to take care of yourself. But a lot of people kind of like that. Florida has a bit of a frontier mentality. Florida is a place where a lot of people move to sort of reinvent themselves. Although it also has a lot of old people. It does. And that was always a little odd for me. And that's one of the reasons we had a high death rate in the end. Yeah. Yeah. And, but of course,

then the DeSantis people would be like, well, if you age adjust for death, we didn't do that badly. Like, well, we still had a lot of deaths here. So, but that, that, the population here was sort of willing on the whole to take the risks and the people moving here were willing on the whole to take the risks. And that's how we wound up in this circumstance. But now you're saying like in Miami-Dade where I live, we're starting to see net migration outflows, the inflationary effects of all these people moving down and then the inherent inflation that occurs. occurred post-COVID is still left at scars here. The cost of housing is really high. The barriers to entry are really high. And there's a good sense of disillusion among a number of people. Mark Caputo, thank you so much for joining us. Sure. We'll be right back with more central air.

Megan, you got to be the main character a little bit this week on Twitter. I did. Not my, not my first time at the rodeo, but yeah. Do you enjoy that? Is that fun? Does it bother you? No, it doesn't bother me, though. It's just sort of a fact of life. Yeah, I mean, this is a funny, like the, because I occasionally have friends who are not in the public eye who then end up getting exposure for one reason or another, and it's an extremely disconcerting experience for them. Yeah, sure, like the first 80 times. Yeah. Or occasionally, you know, I have, I have Twitter set up so that I can only see mentions from people that I follow. So sometimes, I'm the main character. I don't even know it because I can't see any of the conversation. And now I, I, Sometimes people are mad about me on Blue Sky, which I'm not even on.

And so I don't, you know, but occasionally someone will come to me and say, you know, Josh, they're talking about you. And I'm like, I don't, I don't want to know that. I don't care that they're talking about me. One of those people, people happens to be me. When we were at Welcome Fest and you got protested, I was sitting next to you scrolling through Blue Sky where people were calling you a fascist. And I remember looking at you and they're like, hey, you're getting, you're getting really dry. Every tell you on blues guy I have. He looked at me with such contempt.

But I was like, all right, I'll keep my gosh, me. Come back to you. I want to keep updates. But I really, we're the weird ones. Like, it's normal to actually find this to be an extremely psychologically disconcerting experience that, you know, people are sensitive to criticism and are not wired for the sort of society where tens of thousands of people can talk to you directly. And so the experience of receiving a lot of direct negative personal feedback to

to get ourselves to the position where we don't really care makes us quite unusual. Yeah, it did bother me. But by the time you've had someone email you a picture of your house with a gun site superimposed on it. Oh, Jesus. You know, saying that I'm a bad person doesn't even register. Yeah. So what were people mad at you about? AI.

Because I said I use it. Uh-huh. Well, why don't you first describe for us how you use AI? Because I think, I think, you know, if you're not one of these brain poison people, it's still interesting to hear about how this is changing the way that you do your job as a columnist. Yeah. So I use it for a bunch of stuff. Like, starting with when I do a column, I usually, I walk in. I've already got my thesis. I've been reading, you know, but I'll walk in and say, hey, I have a chat bot that's custom. And so I'll walk in and say, like, we're writing about this. Find me. the 10 best papers on this, I want both pro and anti, and then I'll read those papers if I haven't already. I will then write the column, you know, having looked at various research topics, and then I will go back if it's over length, which I'm afraid it usually is, I will say

where are the baggy bits I could cut, I will ask it for a fact check. I will get Claude. Claude is better at this than GPT, so I'll get it to download data sets. And then I will run it through a fact check. I also do my fact checks, and we have human fact checkers. This is just another layer. But I'll say, like, go look for sources. I want links for everything.

I don't take everything it suggests, but I do use some of it. Yeah. And so basically you're using this like a research assistant mostly, it sounds like. Yeah. The people that seemed to respond to you treated you like you were using it to generate your copy, which seems to be the moral, that's a problem. Yeah, it was, it was really weird because I kept saying that's not how I'm using it. And they kept coming back and saying, you're using it to outsource your thinking. This is the thing I've been doing for 25 years. I read stuff and then I write it up. Well, I mean, the other funny thing is like, in terms of outsourcing your thinking, lots of columnists have used human research assistance for a long time. All the columnists at the New York Times get research assistants assign. to them. Sadly, I cannot afford one. Right. But yes. And then more broadly, like, you know, on some level, every part of column development that involves reading other people and talking to

them involves outsourcing some of your thinking in some way. Part of it is that people have outsourced their thinking to other people who say, this is a stochastic parrot, is the favorite phrase. And so they haven't updated. They don't use the latest models. And they're just unaware of what it can do. And they think it's 20, 23, and that it's just hallucinating left and right, which it no longer is. I use the AIs the exact same way that you do. I talk to them constantly and then sometimes I explore my own thoughts, sometimes those things turn to stories and sometimes they don't. But I have to say, like, even though it has gotten better about hallucinations, I still find it wrong all the time.

You know, in various different levels. Yeah, the funniest thing, actually, is when I'm trying to do something on the news, and it will tell me that something that just happened has not happened. Oh, yeah. And it's a prick about it, too. What I was, yeah, I was looking at the Iran invasion right after it happened, and it was like, we have not invaded Iran. It was like, oh, I beg to differ, sir. We most certainly have. Well, I mean, we haven't. Okay, technically not an invasion, but he was saying there was no conflict that I made that up. It's funny. I use this probably less than I should. And it's partly just like, you know, I have a way that I work that I'm used to. And I probably, like a lot of people, I'm not as inclined to update my professional approach as my, as I ought to be. But I don't know. Like I certainly, I don't object to other people going and using these tools. And I'm trying to find, you know, I do use them sometimes to hunt down data and to collate it and that sort of thing is the main way that I've used it professionally. I also use it in personal application. but find the results to be really mixed. In particular, with travel planning, I often find it has poor judgment about where things are located relative to each other or will otherwise make suggestions that are logistically impossible in terms of be here at this time and then be at this

other place at this time, even though it's too far to travel. So one of the things that I take away from that is that, you know, it's good at it's good at digesting texts, but there are certain kinds of knowledge in the world that aren't contained in the text that it hasn't yet gotten good handling. I assume it's going to improve with that over time. It's also, by the way, completely useless if you're trying to buy airline tickets. It can't help you figure out what's available for sale at what price and at the right times,

etc. Which seems like something that actually should be relatively simple for the AI engine. Yeah, I haven't tried it that way. It's not any good for it. And some of it seems to have to do with you can get to something in a web browser, but that doesn't mean it can necessarily access it through the back end. In particular, it doesn't

seem to be able to query the airline reservation systems in the way that you very easily can through the front door. And I assume over time they will find ways around that. But I've just been, I've been interested to see, you know, it's obviously super impressive in certain ways, and it still has deficits in other ways, even though certain things have improved a lot. But I think to the point of the people screaming about it on social media, it's important to be mindful of those limitations. And I think you've described the ways in which you are mindful of them. But I think there's a few things going on here. One is that there are people using AI in stupid ways that, you know,

are outsourcing parts of the job that they really need to do themselves, that are ending up, you know, making factual errors or plagiarizing other people's work. The New York Times just had to fire a book review freelancer who had used AI apparently to generate a book review, and then the AI engine went and stole out of a review of a book that had run in The Guardian. So I think partly people are reacting negatively because there are misuses that are happening out there. And some of it is just this sort of religious anti-AI disposition that exists on the left. that has caused people to form a view and then not actually real-world tested.

That's the stochastic parrot thing. But then the third, and my least charitable take on this, is that, you know, people talk about AI slop, but there's been a lot of slop on the internet well before AI came around. Say it ain't so, Josh. Humans used to have jobs producing the slop. Yeah.

And particularly, you know, in the sort of 2013, 14, 15, 16 era, the height of BuzzFeed listical, and, you know, when the Huff Post newsroom was at its apex, et cetera, you had a lot of these people whose job was to just write the same replacement level takes that were going to go viral on Facebook. This is the Why Breakfast is racist era of web journalism. And those jobs fell apart. Now, first, they fell apart because that revenue model fell apart and you couldn't actually finance a website staffed by humans on ad revenue because the ad rates kept falling. Well, actually, it was more about the algorithms.

Well, yes. Sorry. Facebook changed its algorithm. And strategies that had been very successful stopped being so. Which is hilariously an AI story. Well, I mean, I don't mean to brag, but according to the Wall Street Journal, one of the impetus for changing that algorithm was me at Mother Jones. And they had a little PowerPoint at this thing that was like, Mother Jones has been doing too well at this at gaming this thing.

Tell us your secrets, Ben. How did you game it? I mean, the very short version of it is, is that there was different types of actions that built up affinity. And for a long time, that affinity decided essentially whether it would appear in someone's feed. And there was also a recommendation engine. All of these things would happen by like, if you shared a post, that was a huge,

strong indicator, right? But there's lesser things, like liking a post. That's also a strong one. But clicking on a post, that's also one. and not scrolling past something is also one. And if you can get the click, which is the easiest fucking thing in the world, you're basically then fucked.

You're going to see every post that person does. And so basically, you can have an investigative journalism magazine that writes these serious things suddenly start to grow massively by simply posting free public domain images of space from NASA because someone will click on that beautiful photo and then the next thing they're going to see

is a story about how private prisons are. And so it gets a little complicated to that. But the irony I think of that you brought up that this was such a big thing that it destroyed a lot of these jobs is that that happened right after Trump, right? When there was this sort of moral panic about how Facebook had elected Trump.

And so all of these places bullied them and said like, oh, you did it, you did it, you're evil. A lot of the people at Facebook, who were quite liberal, believed it. And we're like, oh, God, we've done all these things. And so then said, we're going to change it. And they changed the algorithm and killed all those people who had been demanding it. And they should have just shut up.

Yeah. They particularly changed it to disfavor politics coverage. Yeah. They just decided to get out of that business. And it was like an incredible own goal. But the thing is, you know, like you shouldn't use AI to write a column for several reasons. which is why you're not using it to write your column. But one thing about AI-generated writing

is that it's most effective at replacing the lowest-quality human-generated writing. Like, if you ask AI to write a cover letter, it can really do a pretty great job writing a cover letter because a cover letter doesn't say anything. It is one of the most contentless forms of human-generated writing, and therefore the AI can replicate the human there. A lot of the people who are currently or formerly in journalism who are the maddest about AI,

are people who were in the slop business and are therefore their work, if it had not already been undermined by Ben breaking the algorithm, would be automated away by the robots. So in a weird way, it's like they, they in fact are the most AI exposed, but because they're the human version

of a stochastic parent. Like, it's, you know, anyone can say the same things that all of the 27-year-olds at Vox believed, the very narrow space in our politics that was completely dominant. in this part of journalism. And so I, you know, I think a lot of this is guild mentality. The people, people are mad because they feel like the work that you're having the machine do

is work that a human should be paid a salary to do a human like them. Part of the problem is that a lot of the work that they were doing is actually not really better than what the machine can do. By the way, I just got GPT to write me an article on how breakfast is racist. Oh, really? How to do? It's not bad. It's okay. One of the things that I thought was so interesting about those responses to you, Megan, was that there was a certain type of them that said, my God, anyone who has a professional writer job who uses AI at all should give up their spot. You know, they should let someone else have it who's truly going to care.

And like, these people, one, like, that's not really how it works. And the fact that they're looking at it like that, if somebody is better at doing the job that you do and fill the role at Washington Post, they can get the job, but they're not. Like, this isn't a little gift that the Washington Post is doing for writer, charity, writers. Look, I think the fair way to put this, I have seen this before and people screaming in me is that I did get one of the last lifeboats in my industry. And I would be mad, too, if I couldn't, if I watched old. people sailing away to safety, such as it is, well, I could, well, I was just stranded, right?

And, and that's a real, we all feel those sorts of things. It's then amplified by the fact that my politics are usually different from theirs. And I, I do have, I have some empathy. One of them said something quite funny, I thought, which was that they would say, like, give a, get your spot up to someone who loves writing. And like, people love writing congratulations, but like you don't need to love your job. Nick Hornby wrote a book a few years ago called Shakespeare Road for Money, which has everything you need to know is writing the title. Yeah, I mean, I think one of the things that people don't understand is that like, and I've had this conversation with people who are mad in this way before, is like I did this for free for five years before I got paid for it.

I really love writing. I was writing at my peak 15 to 20,000 words a week at the Atlantic. 10 to 20, it would depend on the week. Now, a lot of that was block quoting, but I was generating a phenomenal amount of copy. And that is what it took to get one of those last seats. And I'm sorry, I'm sorry that there aren't enough lifeboats. I would love my industry to be thriving and full of jobs for all these people. It's not their fault. It's not my fault. It's not my fault. It's just reality bites sometimes. It's funny when people talk about loving writing like that. Because as I think about it, this is a question that I haven't gotten in a few years, but I used to get it a lot more.

As people would say to me, you know, I want to be a writer. What should I do? And I would say to them, well, what do you want to write about? And sometimes I would get this blank look back from them. Writing is a skill that some people just don't have and they can have subject matter expertise and they still won't be able to write. But the flip side of that is that, you know, you have to have something interesting and novel to see. And so the advice I would usually give people if they wanted to go into journalism is if there's an industry or, you know, a type of institution that you want to cover, go work in that. Yeah.

You know, go, you know, if you want to cover finance, go work in banking. If you want to, you know, cover government, go get a job in government, you know, in some sector of the arts, whatever it is that you think is interesting. And then you'll develop subject matter expertise and that may put you in a position where you can then enter sideways, where you can run. write about the thing that you know a lot about and do something different. And if that doesn't work for you, you have lined up a career in something else. Yeah. And so that is, you know, that that is the advice that I used to give to people. I think now the, you know, the fact that the, that our industry is in such an economic mess, I think, has discouraged some people. I mean, it's, it's interesting. We had Tyler Austin Harper last week who left a one of the, a tenure track

job. So he, you know, also got one of those lifeboats for himself and then jumped out of it to get into a different one in journalism. A much leakier lifeboat. I mean, and the thing is, like, Tyler's having a very successful launch to his journalistic career. And so I think that, you know, I think he correctly looked at the situation and there were reasons for him to think that he was going to be able to make this work and he has a good job at the Atlantic and he's getting a lot of attention that would not necessarily have applied to anyone in his own way. He did that version of the sideways entry. A lot of his writing has been about academia, which he was indirectly. and then he executed that turn into journalism, which is still available on some scale,

though maybe not as much as it was 10 years ago. The thing about what you're going to write about, a big part of the problem is they all want to write about the same stuff. Right? Like, there are way more opportunities in writing about business or tech or something, but they all want to write about politics and, like, social justice and other stuff where it's, just a really crowded market. So please, please, get expertise in something that doesn't have a zillion other people competing with you for that job. There's also this weird additional thing with that, though, which is like all newsrooms are full of liberals, but like the least politically

liberal part of a journalistic institution typically is the part that actually covers the government. Yes. And I think that's in part because, you know, to cover the government effectively, you have to try to understand people on both sides of the aisle, having a really deep, understanding of policy, you know, like, you know, you may have big idealistic views, but they eventually run smack into reality in certain areas. Whereas, you know, if you're, if you're on the food and cooking team, you can have whatever opinions on politics you want, and the industry tends to attract people who are very liberal. Tech somehow has ended up being covered basically entirely by communists. Or by people who hate their subject. It's really strange to me.

Why are you, why are you spending so much time talking about something? you hate. Yeah. Tim Lee, who writes the Understanding AI newsletter, it's sort of, you know, it opened up a space for him because for whatever reason, the whole section of the, of the journalistic industry has been colonized by these people who just really hate the tech industry. It's also been interesting, though, for me to watch Kevin Ruse at the New York Times. A noble exception. Right. But and also someone who I would have generally characterized as having, you know, quite far left politics. But he, you know, he sees things changing in front of him and has been trying to say like, guys, guys, this stuff is actually. important and does impressive things. And we have to, you know, figure out what to do with that rather than just dismissing the whole thing. But it's been interesting to me how, how unusual he has been in being able to update his priors on that in a way that people around him, if not. The Wired article about acknowledging him is what I feel like I saw people getting mad at him and some other tech reporters who would all said, you know, I used it a little bit. He said he was

using it for a book and people were like, oh, my God, Kevin, I don't love you anymore. How dare you. you're you're Vichy And part of the defense of the people who were mad at me Is that that article did have one guy who was using it to write Just to write his stuff And I thought of myself as critiquing that guy But they perceived me as supporting him Is that just because they can't read?

Like why did they perceive you that way? It should have the AI ingested and summarize it for it. I don't know You guys were talking about the guy from New York Times who was fired a few days ago. Did you read the explanation he gave to an interviewer on Subtack? The freelancer. No, I didn't see what the explanation was. His explanation is he's British and he's talking to an editor from the spectator's substack.

And the guy was like, why would you do that? You're in the New York Times. This is the one time. If you're ever not going to plagiarize and use AI, it's this moment, right? You couldn't write 1,000 words. On a book you hope you read. And the guy says, oh, it was so stupid, you know. I wrote this draft and I was rushing and I gave it to the AI and asked it to change my British

spellings to the American spellings in now style. And then it was under length and I asked if it would help expand and smooth it. And then I didn't check. And I apparently had just. dropped in language from The Guardian. And like the changing the spelling of theater and favorable and stuff is the stupid use. But the second one of expanding and smoothing sounds a lot like a euphemism for I did want it to do a little bit of the writing. Yeah, because you can say where should I expand it?

Yeah. Right? Without saying write it for me. Yeah. Yeah. I think we can leave that there this week. Ben, Megan, thanks for speaking with me. And Megan, thanks for not sending your AI agent to represent here.

Next week. Next week. Central air is created by me, Josh Barrow, and Sarah Fay. We're a production of very serious media. Jennifer Swanick mixed this episode. Our theme music is by Joshua Mosher. Thanks for listening and stay cool up there.