How the tech world is responding to tariff chaos
01:34:42 Share

Support for the show comes from Charles Schwab. At Schwab, how you invest is your choice, not theirs. That's why when it comes to managing your wealth, Schwab gives you more choices. You can invest and trade on your own. Plus, get advice and more comprehensive wealth solutions to help meet your unique needs. With award-winning service, low costs, and transparent advice, you can manage your wealth your way at Schwab. Visit schwab.com to learn more.

Support comes from ServiceNow. We're for people doing the fulfilling work they actually want to do. That's why this ad was written and read by a real person, and not AI. You know what people don't want to do? Boring, busy work. Now with AI agents built into the ServiceNow platform, you can automate millions of repetitive tasks in every corner of your business, IT, HR, and more, so your people can focus on the work that they want to do. That's putting AI agents to work for people.

It's your turn. Visit servicenow.com. At the Home Depot, Spring Black Friday is here, and we've got 14 days of deals to transform your space. So what are you working on? How about a quick and stylish patio furniture update? And what's outdoor dining without a shiny new grill? Find a wide selection of grills under $300, like the Next Grill 4 Burner for only $229 at the Home Depot. Then add a little ambiance with String Lights.

Shop 14 days of deals during Spring Black Friday, now through April 16th at The Home Depot. Hello and welcome to The Verge Cast, the flagship podcast of being nominated for the Webby Award. Yeah. It's good. It's very exciting. David, tell the people what's going on. I thought you were going to say we weren't nominated for the Webby Awards, but we cover the podcasts that were nominated for the Webby Awards. Yeah, it's not us. It's everyone else. The Verge Cast is nominated for Best Technology Podcast, Webby Awards. It's very exciting. Yeah.

We care a lot about this award in particular because it is voted on by people like this. I care so much more about what people listening to this think than, you know, random fusty judges. I like to think of it like the Academy where it's just like a bunch of old white dudes who haven't seen a movie in 50 years. Like, I don't care about them.

vote for us so that we can win and we can crush our competition. Our competition, by the way, is like lovely people, including like the EFF. And I want to destroy them. So we'll put a link in the show notes. I think there's another week to vote. We'll put a link in the show notes. We'll put a link in the container post on the site. But please go vote for us. And I think you can vote for us multiple times. It's like once a day or something you can vote. So like maybe

Make it your homepage and vote for us every single time you think about it, and we will be eternally grateful. Have you been dying for an AI agent project? Here's one. Can you vibe code your way to a Webby victory is a very good question. That's actually a perfect VergeCast episode title. I want to be very clear about that. And that's what we're going to do for the rest of the show. We're going to live vibe code. I've opened Claude.

Uh, David's going to open cursor and then we're going to race. I love it. By the way, I'm your friend Eli. That's David Pierce. Hello. Jake Castronakis is here. Hey buddy. Hey, good to be here. All right. We've got basically three lightning rounds. That's how David has structured this. Although I will say this first lightning round is just a segment about tariffs. We're just calling it a lightning round to pretend it's not just a full segment about tariffs. Yeah, pretty much. There's just, it's like, it's, it's tariffs is one of those things that it is. It contains multitudes.

It's true. Lightning round is like, let's just talk about it all very fast. Lightning is the speed at which these tariff changes have come this week. I'm afraid to say a single number on this podcast because we're recording Thursday afternoon. It's going to come out Friday morning. Anything could happen in those 12 hours. It's so true. My favorite thing about tariffs, and this is very inside baseball editor stuff, is we often assign stories. They're like, here's how something happened, right? Like,

Here's the process by which Go90 went 90. And like, we talked to all the insiders and like, this happened on this day. This happened on the next day. Here's the big decision they made. You see these stories everywhere. Whenever a big decision is made or there's a deal that's done, someone gets what's called the tick tock.

And it's just like, here's all the stuff that happened in order. And those are really fun to read. They're really fun to write. The funniest thing about the tariff situation is that all the major newspapers tried to do a TikTok story of what is by all accounts, pure chaos. And so they tried to impose this structure on like, then the treasury secretary did drugs and called all of his friends and his friends were mad. And they called Susie Wiles. And you're like, this isn't a process story.

Because there's no process. Right. But that's very much like the, like, human beings have a desire for it to make sense. And this refuses to make sense. Charlie Wurzel, our friend over at The Atlantic, wrote a thing earlier this week comparing kind of the response and conversation around all of this to QAnon, which I love. Right. And it is exactly that thing. It is this relentless belief that,

Yeah. Yeah.

I think it was Bill Ackman had a tweet that was like, this is the art of the deal. And it's like, bro, what deal? Like no deal was made. I like, I'm not even arguing that it was a good deal or a bad deal. I'm asking you to identify the deal. Yeah. Okay. But because it is lightning round, the conceit of a lightning round is that it's like lots of different things jumbled up and we just move through them. So I'm going to start the show somewhere else. It's not tariffs. And we can see if we can get to tariffs.

You with me? Okay. Just a gesture of this being a life. Oh, interesting. Because I put this here at the top for you because this is the most important news in the world if you're Nilay Patel. And so this is where we start. But if you can walk me from here to tariffs, I will be very impressed. I think I can get there. Okay. So as you know, I divide the eras of human existence into

Into eras of Sony providing extra base to its speakers. So I was born of history. I was born in the mega base era, which I think we all recall is when America was great. The eighties and early nineties, right? The mega base button was on the Walkman. We were riding high. Top gun had just come out. You all know what I'm talking about, but we've, you've seen the memes. That's what, I mean, that's, that was it. That was the peak.

We were all young in the mid 2000s. Sony moved to the extra base era, which I believe tracks with our decline. It was really weird. You get, you get rid of this thing. People love you. You add a button called extra base. No one knows what it means. It's going away. They've abandoned that into the ULT power sound era. So that's where we are now. Do we left mega base behind? We're in ULT power. So we've talked about this a lot. I'm in the news.

Is that there are now three new party speakers from Sony, which is further evidence that the market for giant party speakers with led lights around the drivers is so huge that Sony is continually investing in new ones as are all of its competitors. And this market is bigger than anyone can see. You know how they're like, we're tracking this outbreak, but it's much bigger because just to base on the, this is what we can just see. This is the tip of the iceberg.

I'm saying the party speaker market is gigantic. It is under-remarked upon. And it's just the evidence that there keep being new ones that let us know it's happening. I'm curious for your experience here. And maybe this just says something about me. But not once have I walked into a friend's home and seen a large party speaker the size of a piece of furniture.

And yet they're like cryptids. Have you seen Bigfoot? Where are they? Where are they going? I don't know. They're everywhere. Is it malls? They're filling up all the empty malls with these things? It's taco shacks. It's weird vendors in malls. It's car dealership. They're everywhere. And I know this because every single company now has a full lineup in sizes ranging from like shelf size to medium sized child.

And they keep revving them. So the new one is a ULT Tower 9. It replaces, by the way, the Sony SRX-SXV900. It has 25 hours of battery playback, a quick charge option that has three hours of playback after only 10 minutes of charging, and is $900. Wow.

It's a tough, like, 10-minute party foul to charge the speaker, but then you get it back for three hours. The ULT Tower 9 offers both ULT 1 mode, which delivers deeper, lower frequency bass, and ULT 2 mode, which provides powerful punchy bass. Who's using ULT 1? Yeah, I don't know, man. Let's be real. I think this suggests that the next era is going to be the one that is deeper, lower frequency, powerful punchy bass. All ULT 3...

is the next phase. It does appear that it also has HDMI in or optical in, so you can connect it to a TV. And it has improved 360 degree LED lighting that can illuminate more floor space. It's $900. It lasts for 25 hours in battery. If you don't need the battery, there's a $750 version that is AC only. But you need the battery. Like you can't party speaker without a battery. This is evidence that the ULT era is not only here, but

But like it's secretly dominant. Like lots of people have ULT buttons in their home. And I know this because Sony keeps investing in making the speakers and they're getting more extravagant and more expensive. They're moving up market. It's true. It's not no one wants them. We have to make them cheaper. It's everyone wants them. They cost $900. They can last for a full day and illuminate your home. What if it was taller? I don't know.

I just want to say before we move on and you're stalling walking us to tariffs from here. But I just want to talk. I encourage everyone to click on this story and look at the pictures because these things are so large that photographs of them look like the speaker has been photoshopped into some other scene in such a large way that it can't possibly make sense. It really is like it's it's the size of like a 10 year old.

And they're just stuck in here. It's like an alien form just being photoshopped into this photo of a concert. The other very funny thing about these photos is Sony is by far the most restrained in design when it comes to these products. And they still look bananas. Yeah. It's very good. All right. So here, that's the ULT Tower 9, which has both ULT 1 and ULT 2 modes. This is where we are in our ULT era. Eventually, there's something else will come.

But I'm telling you, I was born in the megabase era, and I yearn for the simple pleasures of my youth. And instead, I have the Yuletide Tower 9. The reason I bring this up is because it doesn't appear that Sony increased the prices of these products in response to tariffs. Still $900, but now you get two ULT modes and a wider floor illumination. But last week, we talked about the Bravia 2.2.

Sony's new TVs and the Bravia 8 II. Incredible names. I don't know if you saw this, by the way. We got a bunch of emails from people who were like, I straight up didn't believe you that it was called the 2 II. So I went and looked and I would not like to apologize for not believing you about this stupid name. Yeah. One person was like, it's obviously the Bravia II Mark II. And I'm like, no, it's not. It's just the 2 II. It's the Bravia 2 II. So the ULT speakers, apparently inflation proof, recession proof, tariff proof, same price.

TVs, not so much. So it appears that the Bravia 8 II, the high-end OLED, which replaces the A95L in a perfect naming scheme, is going to be $500 more in the United States than it was last year. And the price is unchanged in Canada. So you can already see the tariff impact there. The other TVs are also going up.

So Sony is already beginning to bake in some tariff pricing. Samsung, on the other hand, announced pricing on the Frame Pro, and it is just expensive, but it doesn't appear the price has gone up. The 65-inch Frame Pro is $2,200. This is for a TV that still has edge backlighting. They're just lying and saying that it's mini-LED because it's mini-LEDs on the bottom edge. And that I would point out to anyone who is listening, Nilay's argument for...

for the frame TV always ends with you don't want a TV. And now actually what you do want is a $2,200 TV. People buy these things. Yeah. And then they use them to show artwork while they watch TikTok. And now you can do that for $2,200. If you are, I will say I'm very proud of our audience because all of the comments on this are

I thought this was a good TV of how expensive it was. And it turns out to be a shit TV. I've never been happier. The audience has started to figure it out. The reason the Samsung one is so expensive, we think, is because they moved all the ports to a wireless One Connect box. Right. So you just plug in the TV and then you've got this breakout wireless box to plug in your game consoles and everything else you can put up to 30 feet away. And instead of improving the picture quality, they spent all the money on this wireless box. This is a nightmare.

But it doesn't seem, and also I think Samsung didn't have to raise prices because of tariffs, because Frame TV is such a shit panel. It's been pure margin the whole time. They could just eat it. That's my going theory. But we can already see, I got there, David. On the high-end TV where people pay the money for the highest-end Sony TVs no matter what, the A95L never went on sale. It was a $5,000 77-inch TV. It has, as far as I can tell, never been on sale.

Those customers are going to pay the tariff money and Sony's going to charge it to them. Yeah, because they can. I mean, all of that stuff is easier at the top of the market. I mean, this is something we're going to talk a lot about here is that like the price elasticity question here is coming up over and over and over in tariffs. And it turns out it's actually easier to raise the price on something that's already $5,000. That has perfect demand, right? Sony sold every one of those in NFLs at full price to somebody who wanted them.

So w w I got here. I said, I walked my way from you LT power, power nines to tariffs. I'm proud of you, David, your challenge in th on Thursday afternoon is tell us what's going on with tariffs in a way that is relevant when people listen to this on Friday morning. Okay.

Tariffs both do and do not exist at all times. And that is the state of uncertainty tariffs. Yeah. So I think where we are as of right now is, uh, we talked a bunch last week about this, this nonsensical set of quote unquote reciprocal tariffs that were not reciprocal tariffs that were based on trade deficits and nonsense equations that had nothing to do with anything and appear to have been invented by various AI bots. Uh,

set the stock market on fire for several days. And then Trump, after saying he got phone calls from 75 different countries that he declined to name, put a 90 day pause on all of it, except for two things. One, he has continued to do an escalating tariff war with China to the point where now all the numbers are over 100 percent and just

We've entered like full fake number territory on all of this, like 50,000%. Like, who cares? Yeah, they're just bigger. The number, it's six blades. That's what we're doing. It's six. Yeah, exactly. Can I just quickly explain the over 100%? Because people have been confused and I kind of get it. Yes, but let me just say the one other thing that's happening and then we can dive in. Because I think this actually gets lost in a lot of the back and forthness. The one thing that was left in place was a 10% base tariff on everybody, which is...

seems simple, but it has become vastly complicated because countries like Mexico and Canada, where they have been exempted from some of these tariff deals are now suddenly like, wait, does this apply to us too? And it appears that it does. And so we are, we are back in this place of, uh, it's, it's not as bad as many people thought it might be at least for the next 90 days, but it's still very bad in the most important country, uh, in this case, which is China. And it is still impossibly confusing to everybody. Yeah. So

So that's where we are now. Explain the over 100% thing. Well, it just means if you bring in something for $10, you have to pay $14 of tariffs. It's just fully ridiculous. Like that's the end of people importing things here. Mr. Beast was pointing out that it will be cheaper for him to make Feastables and his other products outside the country to ship to other countries. But he's going to move production out of the United States because other markets will be cheaper to manufacture and distribute to to get away from all this tariff noise.

That's nuts. Like it's just fully nuts to break the world trade system in that way. You can go read the like we tried to explain the decision. But the answer is like the world financial system started to teeter and Trump blinked and he was like, screw it, 10 percent tariffs. And then there's the underlying question, which I think is one of those interesting political realignment things. The power to implement the tariffs is basically Trump declared a national emergency. Right.

And he was like, no, with the power of emergency, I command you to pay me an extra 104%. And the emergency is the trade deficit in many of these cases, which is not an emergency because that has existed before.

For 50 years and is also not what Trump thinks it is like what the definition of a trade deficit that leads you to do what Trump is doing is not what a trade deficit actually is. And I am not an economist, but I am confident that I am smarter about this than our current president. So the really interesting thing about that is a bunch of hard right free traders.

Rand Paul is screaming about the trade deficit not being what Trump thinks it is. I thought you were going to say Dave Portnoy. Dave Portnoy. I don't think he's a hard right for each. I think that dude is just trying to make some money. But, you know, his number went down, number went up. Like he's reacting to it in real time. But the Rand Pauls of the world are out there giving strong speeches and strong quotes being like, this makes no sense. And then a bunch of conservative legal groups are filing lawsuits being like, this isn't an emergency. You don't have this power.

Which is a thing that you would expect really conservative legal groups to do. Say, like, actually the government shouldn't have this power. So you just see the realignment, right? That's not the political outcome you would expect usually from Trump stuff. We'll see where all of it goes. But in the meantime, the tariffs haven't actually –

been rescinded. They've been paused for 90 days. They're still in massive effect against China, which has enormous consequences across the board. And then it's confusing the

how they apply to everyone else. Well, and Trump continues to say during that pause that what he wants is, I was going to say implied, but he's not even implying it. He's just saying it out loud. He wants people to like bribe him and make deals with him in some meaningful way. Like he, the goal is to get everyone to come and tell him how terrific he is. And that's, that's what he says over and over. And there were reports of, you know, a bunch of

leaders flying to Mar-a-Lago to basically tell Trump to stop doing this. Like Jamie Dimon apparently went and was giving a whole speech to Trump about why tariffs are bad. And Trump thought that was really great. And so it's like it's very clear that the leverage he is going for here remains the same. And the way you do that is by putting a pause on it. Sure. Everyone wants out. Like Tim Cook is going to show up and say, give Apple it out the way that you did for me during your first administration when you went into a trade war with China.

It's true. I think he wants the deals. He wants to be flattered. At one point, he had Charles Schwab and he pointed at him and he was like, look, it's a guy. It's like an actual guy. Which is incredible television. Like all of this is nonsensical. There's that goal, right? The weird, corrupt, let's do gangster deals on the side. Sure. That's Trump. Then there's this other goal, which is also an echo of his first administration goal.

where Commerce Secretary Howard Lutnick is saying, I want factories in the United States. I want the army of people with little screwdrivers building iPhones. I want them to come here. Karen Levitt, the press secretary, was asked very directly by Maggie Haberman, does Trump believe we can make iPhones in the United States? And she said, absolutely. We have the workforce. We have the talent. That has been the goal the whole time, right? There's the little goal, which is

Let's do shady deals on the side. And then there's this big goal, which is like, let's restructure the world and build the iPhone in the United States because that will make us great again. We've been covering that for a long time. Like there's a part of this for me. And I don't know if you two feel the same where it's just a bunch of people grew up and they're having the same experiences that we had 15 years ago. And one of those experiences is Donald Trump insisting the iPhone can be built in the United States.

And then we have to have Obama insisting the iPhone can be built in the United States. And we have to like go through the conversation again. I mean, they keep trying to bring like different pieces of Apple manufacturing into the U.S. And each time it becomes like a little bit of a disaster in some way. Like they were going to try to do a bunch of glass in the U.S. And like, I think that factory just like didn't work out for them. No, they do the glass here. Corning does the glass here. They do the glass. I thought there was something for the Apple watch where they're going to do a specific type of glass and they were building a specific factory. The Corning is here.

But there's that. There's the, you know, they did eventually, they relaunched the Mac factory for Trump himself. But all these things take time, like a surprisingly long amount of time. And we have, you know, TSMC opening factory in Arizona that's just like, you know, we're hoping it can produce enough chips to be useful for basically anybody. And we were hoping that those chips are going to be on a modern enough node that we actually want to use them in our latest gadgets. And so it's just like,

there are these baby steps happening. But even when we put a concerted effort into it, we're never quite getting there. There is the ongoing question of why is this the goal that I have continued to find really fascinating? And I think seeing the way it's being covered, particularly by the kind of right-leaning press, is really fascinating. Because on the one hand, I think the reasonable people that I've

That I have seen make this sort of national security argument, right? That it is, this is a way to control your own destiny. If you don't make the chips, you are reliant on somebody else for something that is like crucially important to modern life.

Sure. I think it is vastly more complicated than that, but I at least like understand the bones of that argument. But we've so quickly devolved now into this place where everybody is saying that factories are manly and like, and that actually the problem is, is men are sitting around playing video games, collecting welfare. And what they should actually be doing is working in factories. And that's how you make men, men. And this is like,

This this is what happens when there's no other move. Like, do you remember when at the very end of the election or at the very end of the campaign in like October and November of last year, there was this whole run of stuff on Fox News that it wasn't manly to vote for a woman for president?

This became a real thing and people were like repeating it. And now you have people on Fox News saying the same thing, that tariffs are manly and and that somehow this is the thing you have to appeal to in order to get people to buy into this thing that just transparently doesn't make any other sense. And so I'm at this place where it's like, OK, even if you want to say it is possible and plausible to move all of this stuff to the United States.

Which it's not. And we should talk about that. Like, it isn't. But there still is not an interesting answer as to why, except that it will somehow return us to the beautiful past that everyone is so excited about. Well, there's implicit in all of that is the only market that matters is America, right? That if Americans make American products for the American citizens, like, we will have an ounce of self-control and respect and we will...

I don't know, once again, smash communism in some way. I don't know. I find that argument very confusing in its way because these are not only low-paid jobs in most of the countries in which they're done, they're increasingly automated jobs. And Lutnik is saying the same thing. He's like, these factories will be automated and jobs like robot technician. And it's like, we're very confused. Like very, very confused about what is actually happening in China and what that ecosystem looks like.

And what the supply chains to make that ecosystem go look like. And the thing that is crazy making to me is this new gloss of like, it's manly to work in a factory is it's still unrelated to the problem, right? It's like, we're having a new, dumber version of a conversation that literally, I think in 2010, Steve jobs had with then president Obama, right? Where he's like, I just don't have the talent in this country. You have to invest in the educational system to make me the engineers to work in the plants. Right.

Not as the people with the screwdrivers, but as the engineering managers making sure the plants operate. You don't have enough. I'm never going to do it. The Mac Pro story is really interesting. There's a really – I don't remember where it was from. We'll find it. There's a really interesting story about that factory getting off the ground. It's run by a company called Flex, which is like a Foxconn competitor. That's how you should think of it. One of these big assemblers.

And Flex is in Texas. That's where they do the Mac Pro. They set up the facility there for Apple, and they could not get screws at Apple specifications in Texas. And the one supplier they had was delivering the screws in the trunk of his car. Wow.

That's the problem. It is not like manliness. It is. The problem is we have not invested for years now in the ecosystem and you go to China and the ecosystem is not about like one company owning the manufacturing chain, like Intel owning the chip fab. It's about lots and lots of manufacturing companies serving lots and lots of customers and they're good at customer service.

And that we just, we have to, you have to like reset everything about our culture to make that go. And of course the administration is in the middle of tearing down all of those other parts also. Like you can't say with a straight face, we want to train people to work in factories while dismantling the Department of Education, right? Like those two things are completely at odds, which

Which, again, there is no plan. There's just not a plan. So I don't want you to take my word for it or take our word for it. You can go read in the it's the Walter Isaacson, Steve Jobs biography, which is mostly a bad book. It has like good scenes in it. The story about Obama and Jobs and Jobs is like, this is never going to happen. But years later, in 2017.

Tim Cook on stage at a Fortune conference gave the same answer about why they can't build the iPhone in the United States. We just run the tape. We just run the tape. The truth is China stopped being the low labor cost country many years ago. And that is not the reason to come to China from a supply point of view. The reason is because of the skill.

And the quantity of skill in one location and the type of skill it is. Like the products we do require really advanced tooling. And the precision that you have to have in tooling and working with the materials that we do are state of the art. And the tooling skill is very deep here.

You know, in the U.S., you could have a meeting of tooling engineers, and I'm not sure we could fill the room. In China, you could fill multiple football fields. That's brutal. That's rough. First of all, we should find out where the meetings of tooling engineers in the United States are and how big they are. Like, what's up, guys? The whole country is depending on you.

And they fit in like the back room of a Denny's. They're like, the pressure is out of control. I mean, that's the problem. If you want it, you have to invest in it. And no company is going to look at this tariff chaos and say, okay,

This is a good investment based on expected return. I need to avoid these tariffs. So I'm going to start training multiple football fields of tooling engineers so that I can build one product. They're just going to find another way to do it. Well, and the uncertainty is so much a part of that, too, that right. Like, I think you could you could make a much more compelling version of that argument if there was a coherent platform.

plan. Even if you disagree with the plan, it was like, here is the future as it is laid out, and we are not going to go away from it. But what I keep hearing from folks and what lots of folks on our team have been reporting about is just the uncertainty is what kills you. You don't actually even know where to invest your time and energy and money right now because the tariffs might go away. They might change. They might move. They might just suddenly evaporate all of your possibilities in one country or another. And

Everyone is just frozen by it. So the idea that you're going to redirect investment actually becomes harder in that. Yeah, I think there's also like even if you know this is going to last for four years, can you even do it within that time to a reasonable scale that will be valuable for you? And that's not clear. And then additionally, like to your point with uncertainty, we don't even know that it would last beyond Trump, given that everybody seems to hate this. There's one person in America who likes Trump.

tariffs and we just happened to elect a president. I'm not sure, by the way, if what Tim Cook was saying there was a football field densely packed with tooling engineers or like a football stadium. I imagined the former. Because if it's a football stadium, you're like, you're doing the math. Like, okay, that's between like maybe 75 and 100,000 people per stadium. You know, your biggest college football stadium is like 100,000 people. So you're like, okay, it's like 300,000 tooling engineers.

But if it's one densely packed 100-yard football field, maybe it's way less than that. I've been thinking about this since 2017. I like imagining that what he means actually is just it's three football fields, but it's just 11 on 11 on each field. All he means is it's actually six football teams. But he said it weird. But it's the full 53. Right, exactly. Look, that is from 2017.

That's not today. That's not Tim Cook disagreeing with Donald Trump today. That is years ago. And even that itself is seven years after Steve Jobs made the same exact argument to Obama. Right. And all that it has done since then is accelerate.

It hasn't gotten better. We haven't invested more into these kinds of roles and this kind of training. And by all rights, the only party that can actually do that is the government because the government is not expecting that to return on investment in the form of a product tomorrow. And so, yes, we should build the iPhone here.

We all did this round of coverage in 2012 and 2013. We did it again when Trump got elected the first time in 16 and 17. Now we're doing it a third time. Only all of us have to be like, no, we are real men, which is deeply confusing. All of my cars are faster than yours. I'm just saying it out loud. One is real loud. Yeah.

It's ridiculous because the problem is the same. The problem is you need apparently six football teams worth of tooling engineers. And I just I don't like how what's the plan there? Because if you want to, it's you can't just punish people into training engineers in the United States. Right. And it's not at all clear that we're going to do anything to make that possible either. But we'll decide in 75 days. Sorry, 90 days. That's right. They're not even permanent. It's like the pause is just 90 days.

So we're already seeing a bunch of reactions to this. Apple shipped 600 tons of iPhones on planes to beat the tariffs. Anything that was already on a ship that was like on route beat the tariffs. So there's a lot of that going on. So I don't think we're going to see immediate changes.

The Sheeans and Timus of the world are in real trouble because the de minimis exception that basically those companies exist is going away. We don't know what's going to happen. We know that hall creators like fashion hall creators are just buying everything on Timu right now.

which is getting ready. Amazon has a weird team of copycat. They're already changing where stuff is coming from. We sort of asked around. You can see other changes. Tobias Butler from Toonshine, who was on the show last week, he emailed us. He said, so with the new 104% tariff, the cost of each manufacturer Toonshine goes up 28%, which will cost $90 each, which gets pretty close to the figure he called last week as this would kill me.

So yes, he would have to raise price from one 99 to two 55 and it's still not worthwhile to switch to us suppliers. So he's the like little creators are like coming to these existential points. He says, I'm not sure what my immediate plan is except to reach out to the community and see what people think about $250.

Framework, which we cover a lot, they've just stopped selling their cheapest laptops here because of tariffs. I think they delayed pre-orders. Yeah, they've gone back and forth in a really fascinating way. And I think actually Framework is a company a lot of other startups look to to see how to do this because Framework spends a long time working on this stuff and is also sort of unusually...

transparent in how all of this works like framework will sell you an individual part of a computer so it just like has to be able to tell more stories about how it makes those things uh but sean hosser wrote a thing for us earlier this week and basically in the in the span of 16 hours uh

tariffs were enacted, uh, framework announced it was going to increase the price over of all of its computers by 10%. Uh, like 45 minutes later, Trump announces the pause 30 minutes after that framework announces that it's bringing the prices back to normal. And then two hours later, it says it's still going to increase prices. And then it, it was not going to

added for pre-orders uh it was not going to open pre-orders on the laptop 12 which is its new cheap one but then it decided to anyway and it's just like this company they just keep basically being like look we have no idea what's going on either we're just going to keep updating this blog post as best we can to tell you what's going on yeah that's right super certain makes sense for everyone who wants to invest in a small company if the company

Doesn't even quite know what's happening. Not just small companies, the biggest companies in the world totally caught off guard. Speaking of updating blog posts, we just have a list of automakers and what they're doing that Andy Hawkins has to keep

because the plans keep changing. It's basically chaos. We will, we'll link to that post, but VW is adding import fees. Stellantis is idling production at various places, putting 900 jobs on hold. Jaguar, Jaguar Land Rover, like the people who live in the town around its factory are basically like, this is the end of us. Ford has so much inventory that they're just cutting prices across the board, which is very good. I,

Audi is just holding 37,000 cars at docks until something happens. Because as long as they're there, they don't have to pay the tariffs. You just leave them. I think they haven't brought them in yet, right? Yeah, they're like- They're like, we got to get these in eventually. They're on the other side of the fence. It's nuts. That's all nuts. And then the main thing that we should talk about is the Switch 2, which-

It feels like there's the weird cultural, like, tariffs make you manly, and then there's the Switch 2. And in terms of just weird reactions, like, the Switch 2, I think, bore the brunt of it. It's also been the most public because Nintendo had the just, like, terrible luck of announcing the Switch 2, I don't know, 30 minutes before the tariff threats began. And so they're doing this press tour at...

as this information was unfolding. And so normally companies can kind of hide from some of these questions if they just don't put their executives out there. But Nintendo is in the middle of putting their executives in front of every single journalist they could find. And all of them only had one question. And it was, what are you going to do about these tariffs? And the price is already high. So there's this assumption that Nintendo

the Nintendo priced the thing high for tariffs. And the answer, and I think Doug Bowser said this to our own Andrew Webster, he's like, no, that was how much we priced it at, and we'll see what happens now. Yeah. Which is not a great answer. But then pre-orders were supposed to have been this week.

And we're not this week. And that was that was the thing that really set everybody on fire was Nintendo went from we're going to sell you are very expensive console that you're going to be slightly mad about, but definitely for sure buy anyway to actually you can't buy it yet because we don't we don't know what it's going to cost. I think preorders for something like this is the worst possible situation to be in because

They're importing millions and millions of these things. And if you're going to sell them at $450 today and you have zero idea what they're going to cost you to bring in two months from now when they actually go on sale, that could just devastate them. And obviously they are going to be

bringing a bunch in early to avoid the tariffs as much as they can. Allegedly, they've been stocking up for months already. But it doesn't really change the fact that they could be put in a position where a bunch of people bought these things early at one price and they actually have to start selling them at a much higher price two months from now. That's just going to lead to a complete disaster for them. Yeah, and there's

there's no good answer at the end of that either, right? Like you, you either immediately increase the price of an already pretty expensive thing. Uh, and you can blame that on whoever you want, but it's going to hurt demand.

Or you say, you know, get it now. Maybe the price will go up, which is weird and bad. It's also for a company that big. It's it's like logistically really complicated to change the price of something. This is something I've heard from a couple of people that I hadn't really thought about before, that like you have to change your marketing materials and you have to change like signs that you've made and you have to change.

like commercials that you've, you've put out in the world that people are going to be running on television. Like it's actually not as simple as just changing the price of a thing on your website. When you want to change the price of something, it actually becomes a real process. And it's, I just like that alone is very complicated, but like, this is that, that exact thing, uh,

like this guy, Dan Soroker, who is the CEO of a company called Limitless. They make one of those like AI voice recorders that listens to you all the time and tells you what interesting stuff you said. I have one upstairs. I haven't used it yet, but

But we'll report back. He tweeted this yesterday. He said, This is the worst SAT question I've ever heard in my entire life. But those are your options, and they're all terrible.

And if you're a Nintendo, maybe you can eat the cost, right? Like if you're Apple with famously high margins, you can eat the cost. None of these companies want to because they like money and that's the job. But a lot of these companies, especially these smaller companies, literally cannot afford to eat that cost. And so now they're like, okay, do I triple the price of my thing or am I out of business? And am I out of business either way?

Well, we're also saying like they might be able to eat the cost. That's like at 10%, they might be able to eat that cost. At like 34%, that becomes a completely different equation. And then at some point, we're talking about these 100 plus percent tariffs. And I think depending on the country that they're manufacturing in,

they're going to have different levels of existential crises here, but all of them are pretty existential. Even 10%, you know, I think we're looking at, that's what some of the stuff that Frameworks is dealing with. Even that is enough to cause them heartburn and go, I don't think we can sell this anymore. Yeah. The interesting question is going to be where the money moves, right?

So, you know, for a product, an AI voice recorder, there's a service component with that. So you might sell the hardware at a loss and then make the subscription price five times higher. Yep. I think we're going to see a lot of those moves, which will suck. Just like flatly that will suck. Yeah. And then we're going to see like the end of discounting. Like most people price in sales like that stuff is going to go away.

So that we don't know what's going to happen. And I think the switch to is going to be a kind of like the leading indicator, how certain anybody feels because it is a product everybody wants and they could probably raise the price and still sell a lot of them. But I don't think they want to. I think they already know it's high and they're already eating all that because it's high.

There's other stuff, weird stuff happening in response, by the way. China is considering showing fewer U.S. films, which would just basically crush the Marvel machine. Right. I mean, less so than a few years ago. Like it's that's shifted a little bit already, but it is still a huge part of the industry. And it's like it was very funny. We were talking about this earlier that like, do you remember how all the bad guys in Mission Impossible movies are now called like the entity? Nobody like fights foreign countries anymore. They're all just like nameless bad guys.

bad guys that's because they wanted these movies to play in other markets mostly China and so like and then China like took all the stuff that it learned from Hollywood built its own now has a huge homegrown movie industry and is now just like yeah get out all that money used to make from showing movies in our country goodbye yeah China has a lot of those moves to make if it wants to it does have a lot of those moves to make I'm just saying that I watched the three body problem and I'm like less worried about it

Than maybe you are. Not so worried about that one. That's fair. Yeah. And not the new Netflix bad one. I mean the Chinese bad one. Just to be clear. The Netflix one was worse. That's like worse evidence. And longer. What are we doing? I'm just over my alcoholism was a real plot point in the Netflix one. All right. We got to take a break. Maybe by the time we're back, we'll know what's going on with Ferris. Let's find out. We'll be right back. Support for the show comes from Charles Schwab.

Support for this show comes from Roomba Robots.

That's right. I said LIDAR.

That's the stuff archaeologists use to discover ancient cities. Roomba robots aren't messing around. The all-new Roomba 205 Combo gives you 60 days of hands-free cleaning thanks to innovative technology that can packs dirt and debris inside itself. And the powerful Roomba Plus 405 Combo comes with advanced spinning dual clean mop pads and an auto wash dock. It vacuums, mops, and then empties its own bin and washes its own mop pads.

You can control the whole system from anywhere with the redesigned Roomba Home app. With all these features at your fingertips, you'll never miss a mess. Roomba robots were made for this. Learn more at iRobot.com. That's I-R-O-B-O-T dot com.

♪♪♪

We're back. I want to issue a correction. It was just called Three Body. It aired on Peacock in the United States. There you go. Also, I just want you to know, people sometimes ask what we did during the break. I went to NewYorkTimes.com because I literally thought it was possible that there was new tariffs news while we were at break. There's not, so that's exciting. Yeah, good. We'll see how this goes in the next 30 minutes. There's tariffs, then there's AI. AI also had, I would say, a messy week, starting with Meta, which...

I think we tried to release a bunch of exciting news about Lama and then all of it quickly became unexciting. What's going on here? Jake, can you explain this one? Yeah. So Zuckerberg, I think he wanted to have a little bit of Elon energy here. So last weekend, I think it was middle of Saturday. He was just like, boom, Lama four. Here it is. Everybody go have fun. And just, you know, kind of try to catch people by surprise.

dropped this big new model that people have been expecting. And, you know, at first glance, everybody was like, wow, very impressive, great benchmarks. And this is the thing. Every single time a new flagship model comes out, the companies brag about these benchmarks and they show how they perform against all the other models. And inevitably, inevitably, their best model beats out all the other models on the market.

And that's what happened here with one of the Lama 4 models. And then eventually people started to look a little bit closer, started to look at some of the fine print on this. And they realized that Meta had actually benchmarked

a secret model that only they have access to, that they're not actually releasing. And so it's like an experimental chat-specific version that is sort of potentially designed to beat these benchmarks.

that they're not releasing to the public. It was just used to look great on benchmarks. And so people have been freaking out about this. The specific benchmark that they nearly topped the leaderboard on was like, hey guys, that's not how it's supposed to work. And Meta was just like,

We have lots of great models. I don't know what to tell you. And so I think it has not been the reception they were hoping for for Llama 4, which is supposed to be this big release, supposed to get them back in the conversation. Meta has been trying to make Llama the big, quote unquote, allegedly open source option to get a lot of big adoption here. And, you know, I think

If they wanted to curry favor with the community, this is sort of off to a bad start because now they're starting to feel a little bit misled. And they're already also starting to, I think, complain about some of the restrictions and limitations around Meta's, you know, so-called open source approach to it anyway. I really genuinely respect Meta's inability to help itself from doing stuff like this. Like,

You go way back and, you know, Meta got in all this trouble for massively inflating the view count of all the videos so that publishers would make videos and like screwed up a whole generation of publishers on the Internet as a result.

This company just cannot help itself from making up numbers about things in order to look impressive. I kind of love it. I love this. Like, let me be clear. Don't do this. This is bad. But also, it really just shows how nonsensical these benchmark charts are. Like, what does it mean that it gets 98% of some random benchmark? Like, I don't know what that does. I don't know what that means. Number just goes up. Yeah.

And I think it's just increasingly clear that these do not actually correspond to does the model give you a correct, accurate, good and useful output? It corresponds to can this model beat this computer test? And in this case, they specifically designed it so that it could beat this test a little bit better than the other models.

Isn't this particular benchmark, doesn't it have humans in a loop? It does. This is a really interesting one where they pair it against output from other models. And so you choose which one is better. And so they're specifically tuning it theoretically so that humans will respond better to their thing. Right. That's the best part of this. It's not designed to be better. It's designed to...

Do whatever it takes to make the humans think that it's better, which is so funny. It's like we don't have to build a better model. We just have to build a model that convinces the idiots that it's better. And then we've done it. Perfect. That's one of the most meta instincts of all time. Like, we don't have to be sincere. We just have to be convincing. Yeah.

By the way, Meta loves to game benchmarks. Like, there's a whole thing here about companies gaming benchmarks in the history of tech. Like, every GPU company has been caught gaming benchmarks. It's just a thing they all do. Every laptop company has been caught gaming benchmarks in one way or another. Intel has been caught gaming benchmarks. But this one is, like, particularly Meta in that they inflated their own video numbers for years.

And then they kind of admitted it and they're kind of like, but whatever, like number was so big. Right. And that's like very much the attitude they're taking with this. I agree with you, Jake. At the end of the day, what really matters is can anyone build useful products out of this stuff? Like these benchmarks do not matter. Like every week there's a new one and another thing is slightly better than the last one. And we're still all looking at fundamentally the same chatbot products with the same distribution. Yeah.

Maybe Meta is going to do the glasses like we've heard about them that, you know, that will obviously rely on Lama in some huge way. But I don't think that matters if it's Lama in there or ChashiBT. Like, it's really the interface and the distribution that, like, carries you forward, not the capability of the model. I agree with you in everything.

Every way except one. And I think the one is that I think we're still early enough in this industry that momentum is really important. And what is definitely true is that LM Arena, which is the website that runs these benchmarks, is

is a website people who make AI stuff look at a lot. And so being at the top of that leaderboard is a signifier of stuff to the people you want to like come work for your company to build cool AI products or come, you know, build with your API and pay you a bunch of money to use it. Like these things are not mandatory.

meant for regular people to care about, generally speaking, they're meant for like the people in the business to care about. And I think that's all the more reason to try and game it right. Like, these benchmarks exist to try and get AI researchers to want to work at meta and not open AI. And

coming out like you're cheating is a tough look. And I think there's been a lot of that forever, right? Do you guys remember in the early days of this, there was this whole big kerfuffle with Apple where Apple wasn't letting its researchers publish publicly their research. And so it was having a hard time getting researchers because one of the things these people want to do is like publish their work and share it with the world. And so Apple has had to rethink the way that it talks about its AI stuff in order to get good people to come work there. And so I think in that sense,

playing these games really does matter, but it's in a much less like human relevant way than like actually making good products, which none of these companies are doing. No, I think that makes sense. And, you know, I think that community is, that community is the people who freaked out about this. They're the ones who feel betrayed because they're, you know, they were excited about this. And I think, uh, meta has made big promises about, um,

you know, how you'll be able to use Lama because of its, you know, allegedly open source approach. It's the opposite of DeepSeek. You just made me realize that. DeepSeek was the one where everybody was like, it can't possibly be this good and then started using it. And they're like, oh my God, it's incredible. This is the one everybody's like, whoa, it's really good. And then it's like, oh, nevermind. It's not that good. You know, it's funny you mentioned Apple there because what I was going to say is, you know who's not on the Alamarina boards? It's our friends in Cupertino. Yeah.

I mean, they have some models. They're not building their frontier model at scale. Big story in the information this week about Siri, actually, at Apple. It's a lot. There's a lot going on in this story. It's basically, here's why Siri is bad. Which, keeping with the theme of this episode, is a story we have been telling for 15 years. But this one in particular has a lot of

Details, some of them shakier than others about what actually happened. Here are the three that just jumped out to me. And you tell me, I know you guys read that story too. You tell me if these are the ones that jumped out to you. One, we've talked a lot about the WWDC video of what like agentic Siri could do on the iPhone with app intents. And you just ask it where your grandmother is and it like finds her and sends her a note or whatever, whatever was happening in the demo. The Siri team, according to this information report, had not seen any of that working.

So that was just a pure demo. And the Siri team was surprised at what they were seeing. That's really bad. That's real bad. Second, I'm just like a decoder org chart, you know, aficionado. John Gianandrea ran his own AI team. He was apparently a very sleepy manager, didn't have a lot of fire. And Craig Federighi built his own team.

to do stuff with machine learning inside of Apple that got bigger and bigger and bigger and started competing for researchers and engineers with the main AI team, which led to him taking over Siri in the end. Which is very bad, right? That's not good news. And then the last one, which is particularly funny, is there was apparently a big debate in the aftermath of the WWDC demo about how many models Apple would use.

If they would have a local model on the phone that responded to local stuff like timers called Minnie Mouse and then a big model in the cloud for big stuff called Mighty Mouse, or whether it should all be Mighty Mouse. And I think they landed on all Mighty Mouse and none of that shipped and none of it worked. And who knows what's going to happen? They picked wrong. I really, like, really, I think the, the, one of the lessons we are learning here very quickly is that actually, uh,

single behemoth models that do everything have their place, but ultimately the answer is lots of models for lots of things. And Apple seems to have just... It's like picking the Vision Pro instead of picking smart glasses, right? Like it tried to do the whole thing and...

in so doing prevented itself from doing the thing that actually works. I'm also really curious about, you know, they took this, we're going to do it. We're very heavily lean on local processing for the AI approach. Um, and I think, you know, if you go back, what, like two years on the pixel, they were like, Hey, you know, you have to have the highest end pixel in order to use Gemini because you need the local processing abilities. You need the Ram. Um,

And then fast forward, like, I don't know, six months, and all of those abilities are just on all the phones anyway, because they're all just doing it in the cloud. And so I'm kind of just wondering also if Apple's still just picking wrong by trying to do so much on the iPhone, because it is sort of against their constitution to offload this stuff to the cloud because of the privacy concerns around that.

Which, you know, that's fair. It's a really great perk to be able to say, hey, this is all local. It's all secure. It's all safe. But it's just one of those things where sort of a self-enforced limitation that's continuing to hold them back. Totally. And the way that they're getting around that when they have to go to the cloud is just by punting it on the chat GPT, which is actually the best way to do it, unfortunately. I mean, probably. But it sure doesn't make Apple look very good.

Okay, wait, while we're talking about org charts, we should move on from this. There's other stuff to talk about, but I just I want to read a paragraph to you guys from this story. It's by Wayne Mott, the information. Shout out to Wayne. It's a good story. And I want you guys to tell me whether this sounds like people want AI or that they don't want AI.

It says, former Apple employees have referred to Siri as a, quote, hot potato, continuously passed between different teams, including those led by Apple's services chief, Eddie Q, and by Federighi. However, none of these reorganizations led to significant improvements in Siri's performance. There's one way to look at this that says, everybody is trying to get it, but nobody can pull it off. And there's another way of looking at it that says, Siri is poison and no one wants to be in charge of it. And I'm starting to wonder if it's the second thing. It's the second. I don't know. I read that same thing and I...

It's funny that you saw ambiguity in it. And I was like, oh, it's poison. Okay. Okay. I mean, that was my initial read too. But then I'm like, okay, there is another way you could look at this. That is like, it's, it's, it's very competitive. Everybody wants to be in charge. I think you're right. I think it's poison. Well, I think there's, you can intellectually make the argument that being in charge of Siri right now at Apple makes you the most powerful person at Apple.

Because all of the things that might kill the iPhone are things that look like what Siri should have been doing 10 years ago. Yeah. Like OpenAI has this project with Johnny Ive for a screenless device. That's obviously some sort of like Siri puck. Right. And maybe it'll work. Maybe it won't. But like, that's the thing. The Metaglass, as we've talked about a lot, that's going to be a voice activated assistant thing. Samsung is going to release the Bali robot. Yeah.

Let's just pretend it's in the mix here. What you want is a ball that rolls around your house and projects Samsung's fast TV service on the wall in response to your voice. Hey, no other AI product has that kind of physical presence. It's true. It's very good. None of them even have a shape. Do you want a small rolling ball in your house that shows up and commands you to exercise on command? The ball is there for you. I do. Like, I earnestly do.

Every Bali demo is like at the end of it, it's like, and now do jumping jacks. It's very good. All of this is predicated on voice working. Alexa is predicated on voice working. So anything that might kill the iPhone has this one interface idea embedded in it, which is natural language. Voice commands will work and the computer will have natural language voice output. It will, it will, it will work.

None of it works, to be clear. Like, Humane existed and then went 90. It famously no longer exists. None of it works yet. But if you're Apple, being in charge of the Siri team is being in charge of the thing that might kill you. And I think they can't figure it out because they also can't break it. I think one of the hardest things about Siri is everyone knows that voice is a thing that might disrupt the touchscreen, which I don't necessarily agree with, but that's what everyone thinks.

But everyone knows that voice is a thing that might disrupt the touchscreen. And then you've got however many years of Siri existing and doing timers and starting music and pretty much doing those things that you cannot break. Right. And Apple talks all the time about how often Siri gets used. And even if it's just for those two things, it is still, it's an entrenched user behavior now that you can't screw up, even though they are doing their very best to screw it up. Yeah. So I just see it as here's this thing where if you get it,

The stakes are so high. Your opportunity to make change is so limited because you can't break it. And then you also have to invent the cutting edge of AI. What you're describing is just what Google did with Gemini already, right? And the advantage they have is that they have a functional AI division. But they had Google Assistant, which was like roughly as good as Siri. And it set timers. It added stuff to your calendar. And when they launched Gemini, it broke all of that. Couldn't do any of that.

And then they've just slowly hacked each one of those things on to Gemini. And now they're just like, we're done. Forget Google Assistant. We had 10 years of brand equity, whatever. Nobody cares. Gemini's new thing. And they just like they're just fundamentally wasn't that much good stuff.

about these old era of assistance that you can kind of just start from scratch and rebuild timers because as much as it pains us that it takes like five years to add a multiple timer feature to each of these things, like that surely isn't the hardest part of building AI. And so we've seen another company do this already. And it feels like what Siri needs is that start from scratch moment. And theoretically they had that. And it's just like not quite clear why that isn't clicking.

Well, two things about that. One, Google had its own bloody org chart battle over making a functional AI division. Google also invented most of this technology. Like there might be no other company in tech

as equipped to do what you just described as Google. And even it was a giant mess for Google. Right. And then they did have two products, Assistant and Gemini, coexist side by side in the most Google of fashions. And then I think if you go and look at the Google Home forums and the Google Assistant forums, the switch to Gemini just being their

Is it going about as well as Google Home goes? You know what I mean? Yeah. Sure. Like, there's enough people on those forums who believe that Google Home is effectively abandoned that it should really worry Google. Right? So, like, yeah, they're in the middle of this transition and maybe it will all come out in the end as something.

sunshine and daisies. But right now it's like, oh, the thing that we depend on feels like abandoned work. For example, this is just a really dumb one. We have Google homes in our house and they are meant to control our Sonos because we have Sonos speakers from the time when those two companies got along and they just stopped working. You could be like, yo, play me the music. And it's like, I'm doing, I'm doing it on the kitchen. And you're like, no, you're not do it again. And it's like, I'm super doing, I'm playing this music on kitchen and it's not.

And there's just, you know, you look at the Reddit post and it's like 400 days ago, someone said that Google Home stuff worked. And they just don't care. And I think that's, I agree with you, Jake. Google has done a better job. Like Amazon hasn't pulled this off with Alexa. If I can defend Google Home for one second, it's been bad for years. Fair enough. Other AI stuff. Jake, you want to talk about the Shopify CEO saying that no one can make a new hire at the company? Yeah.

without proof that AI can't do a job. And you were like, this is what reasonable people think. And I'm excited for you to defend this take. Wow. Sorry. Sorry. Oh boy. Oh boy. Jake, here's this smoking stick of dynamite. What would you like to do? Uh, that's, that's, yeah, that was, is in fact part of what I said. So, uh, Shopify, not, not Spotify. Let's be clear with that. Shopify CEO, uh, Toby Lutke, if I'm pronouncing that, ballpark, uh,

You know, this this memo got leaked. And so he just posted the entire thing on X. You can go and read it. And, you know, the headline of it is, you know, Shopify CEO says, like, you can't hire anybody whose job can be done by AI. And you know what? It is true. He says that in there. Yeah.

But I do think, you know, if you read the entire thing, it is both more reasonable and I think less reasonable than you think. Because on one hand, he's just like, actually, we just all sort of need to learn AI. And the other hand, he's like, we will 100X our productivity. So, you know, I think one thing that occurs to me is that the people who are using AI the most right now, it is the engineers. It is like the developers who are just co-op.

coding all day long using these things. Separately, Anthropic just launched a $200 a month tier this week where they're just like, yeah, we think that engineers will just pay for this when their work won't so they can just like use it all day long. So this is our, I think when you send this memo out to an organization that is substantially built out of engineers, this is not necessarily as scary as it would be if you send it out to your news organization, which would be like, sorry,

And so I think, you know, there's parts of this memory says, here's just some of the preface where he goes, quote, what we have learned so far is that using AI well is a skill that needs to be carefully learned by using it a lot. And it's like, okay, yeah, this is actually like, this is like pretty, pretty reasonable. You actually should test these tools to see if there are ways to build them into your workflow. But at the same time, he is saying, I think the very big, scary thing that has everybody worried, which is,

yeah, these tools are going to number one, change how we work. And in doing so, start to potentially replace roles that would otherwise have been hired. And so I think this is, you know,

This is a very scary memo. I think it's also just worth reading and understanding because this is actually just how a lot of these companies are going to be thinking about this. And I think, is AI ready to replace a single human right now? I sure wouldn't make that bet. Is it ready to do little bits and pieces of some people's jobs? I think the answer is clearly yes.

especially if you look to developers. And I think, you know, we have an engineering heavy organization here that is saying, hey, let's start using this more and more to get a little bit more efficient. Some of the stuff he's saying is like, actually, you should just use this in your prototypes, right? It's not necessarily you should go and replace your coworker with it. He's just like, it's like, I'm making it mandatory that you use this in your prototyping phase. Okay.

It's funny because I didn't read that as being about the engineers. I read that as being about product managers and salespeople. And people have ideas for new features. And it's like, you're just going to vibe code out a garbage version of the new feature so you can push the buttons and play with it and see if it's any good. And there's something about that that really speaks to me. Because I can't do that. And this is the thing I keep thinking about is...

Our own company is like putting out ads being like, stop AI theft government. Disclosure of Box Media is putting out an ad with a bunch of other publishers being like, stop AI theft government. And, you know, our team has thoughts about AI and I certainly have thoughts about AI. I think I'm a better writer than AI to this day. Declarative sentences be damned. That's what AI thinks I write like. Yeah.

But I can't code. I certainly couldn't code like a prototype feature of a thing that I wanted. And the idea that this tool can let me do it is super interesting. And then you see even in creative industries, people are complaining about it, but then they use the features like mad. So I would pair the Shopify thing up with what I think is kind of the most interesting AI feature that's been announced in a long time, which is Adobe building AI agents in Photoshop and Premiere Pro.

Where you're just like doing Photoshop work and it's like, do you want to do this thing? And it's like super clippy and then it just does it for you.

like it clicks around the photoshop interface for you and just does it i think there's something super interesting about this photoshop example too where if you watch the video of what it shows that that it's suggesting you do it's stuff like add a blur to the background and it's like three years ago nobody would have called that ai that like google photos just auto suggest that on literally every photo and so like there's just this broad spectrum of what like quote unquote ai is now and

And there are, I think, these small tools that people are just sort of using instinctively every day without quite considering them to be AI because it's just, oh, this is just making me like 2% more efficient. And we sort of freak out when it changes from this 2% thing to the like, I must replace my coworker because my boss won't give me more resources, which is like concerning. I mean, and a funny thing to do is like, if you just...

Yeah.

What's different now is that everybody is telling you that they're building God to do it for you. Well, right. But I mean, like the I don't need the accountant because I have Excel is like one version of this. I I'm a marketing director and I can build a fully functional prototype of a feature that I want to ship in the main software. Very different. But I think the second thing is less existentially frightening than the first thing. I think we won't need accountants anymore because Excel exists forever.

was much more plausibly threatening to people's jobs than now I can make a crappy prototype by myself. It is true, by the way, that Excel got so complicated that we need accountants more than ever. And maybe that's just the trajectory that we're on. Accountants just use Excel now. That's the only thing that changed is accountants use Excel. All right. I don't know on what timeline Google Assistant will be good, but I do know that we should probably put Shopify in charge of it. We got to take a break. We'll be right back.

I was never really a runner. The way I see running is a gift, especially when you have stage four cancer. I'm Anne. I'm running the Boston Marathon presented by Bank of America. I run for Dana-Farber Cancer Institute to give people like me a chance to thrive in life, even more.

Join Bank of America in helping Anne's cause. Give if you can at bfa.com slash support Anne. What would you like the power to do? References to charitable organizations is not an endorsement by Bank of America Corporation. Copyright 2025. At Capella University, you can learn at your own pace with our FlexPath learning format.

Take one or two courses at a time and complete as many as you can in a 12-week billing session. With FlexPath, you can even finish the bachelor's degree you started in 22 months for $20,000. A different future is closer than you think with Capella University. Learn more at capella.edu. Fastest 25% of students. Cost varies by pace, transfer credits, and other factors. Fees apply. This episode is brought to you by Indeed.

When your computer breaks, you don't wait for it to magically start working again. You fix the problem. So why wait to hire the people your company desperately needs? Use Indeed's sponsored jobs to hire top talent fast. And even better, you only pay for results. There's no need to wait. Speed up your hiring with a $75 sponsored job credit at indeed.com slash podcast. Terms and conditions apply. ♪

All right, we're back. David, it's time. Oh my God, it's time. We've waited this long. Once again, we are here for America's first or second favorite podcast within podcast, for which we have now received theme music that I have to figure out if I'm allowed to play on this show sent in by listeners.

Once I know if that's legal for me to do without asking permission. Starting next week, we have theme music. It's time for Brendan Carr is a dummy. It's funny because on this America's favorite podcast within a podcast, except for the other podcasts within a podcast that people love. Top two. Yeah. We've started to receive fan submissions for parts of the segment. Yeah.

Like more so than I think the main Verge cast at this point, which is very good. It's good stuff. So this week, a few things to talk about this week, but the main thing, so many people sent me this photo of Brendan Carr, chair of the FCC, smiling in his dumb suit, wearing a pin that is the gold head of Donald Trump. Just so proud of himself that he's wearing a pin that is the gold head of Donald Trump.

Look, there are some like overheated blog posts about this thing, about how this is Maoism and this is what Mao did in China. Like, Brendan's not that smart. You know, he's not intentionally making a callback to the Maoists in the Cultural Revolution. No. He's just a dummy who wants to impress his buddy Donald Trump and doesn't realize that

That being the head of our nation's communications infrastructure should require him to be nonpartisan. Instead, he's making it plain that, in fact, he is a political flunky who shouldn't have this job by wearing the gold head of Donald Trump on his chest. You don't want that. If you had a Democrat wearing the gold head of Barack Obama in charge of the FCC during the net neutrality battles, I'm pretty sure the Republicans would have burned the building down. Yeah.

You don't want that. You don't want it on either side. You want these people to be relatively neutral. The reason you have a commission that's made up of Democrats and Republicans by statute is so that when they take votes, they're not political. That's not how it's supposed to work. Right. They're supposed to reach some compromise and regulatory affairs. But anyway, you got this dummy wearing the gold head of Donald Trump.

That's that's just that's our boy, Brendan, making it plain that he's just a political flunky. He's there to do the president's bidding and the president's bidding is to chill speech across the board. It's not important, but this pin itself is is pretty spectacular. Like I have to say, it's it's definitely like some BS fake gold stuff. But in it, Trump has substantially better hair than he does in real life, but is also making like a little bit of a kissy face and the direction it's facing on Trump.

our man Brendan's lapel kind of makes it look like Trump is about to like kiss him on his neck and that's just that's just a visual I really appreciate the first time I saw a photo of this I was like this is 100% AI there's just like there's no chance this is real this is too insane and regret to inform you all I was wrong yeah it's a real bummer I can't believe I'm going to say this but

You know, if you pay attention to Trump, he's always kind of making a kissy face. That is true. He does have sort of like resting kissy face. It's really true. He's not lost in thought. He's just pursing his lips. I don't know what's going on there. I want to call out our friends at Ars Technica who wrote a huge profile of Brandon this week. They called him the speech police. They're going through all the news distortion cases that we have covered endlessly on the show. It's just very clear to a lot of people.

That Brendan Carr wants to censor the internet and censor broadcast regulators. It's not just us talking about it. We just have the most fun doing it.

Because my brain is broken. It's out there. It's out in other publications. I think Rolling Stone did a piece. The Ars piece is particularly good in a way that Ars is particularly good all the time. It's also starting to hit the other commissioners in the FCC. So I'm going to do something we don't usually do here on Brennan Carr's Dummy America's Favorite Podcast. I'm going to point out that the other commissioners of the FCC are doing a good job.

So in particular, Ana Gomez, who is a Democratic commissioner, was recently National Association of Broadcasters Convention, the NAB convention. And she directly, openly called it out. She basically said that Brennan is a censor and the FCC is totally out of bounds.

She said it is particularly important for media regulation, which is why the FCC was set up as an independent agency a long time ago, because the fear was the type of interference that you are now seeing today with the White House with media, whether it's broadcasters, public media, Internet platforms in the attempts to control speech.

And then she went on to say, we've seen this administration throughout the administration threaten tech companies for their moderation practices to give consumers an environment that they want to that they want to stop fact checking. It is an absolute pattern of censorship and control. If we let them do this, it will be to the harm of the country. So you have at least one FCC commissioner saying what we're doing here is wrong.

just fully wrong. And we're actually doing the thing the FCC was set up to make sure it didn't happen, which is censor the speech of Americans. So good job, Anna. If you want to come on Dakota, you're more than welcome to. We've extended that invite. I've got one last one on Brendan. This is maybe the single most predictable Brendan Carr is a dummy item in the history of America's favorite podcast podcast. It's a

It's an increasingly high bar, that one. I mean, I'm just going to start to say it, and you will complete my sentences for me. Okay. This week, Brandon put out a blog post announcing the FCC's next open meeting, which is when they go through agenda items. He titled it Spectrum is Back Again because he thinks he's your buddy instead of being a vicious internet censor. And what he's excited about is opening up more spectrum, which is basically the job of the FCC. And they want to open up satellite spectrum that is currently being used by geostationary satellites. Okay.

You can guess. You can guess what's happening here, right? SpaceX filed a petition to open up satellite spectrum that's currently being used by geostationary satellites. And Brendan is taking up so he can open it up without once mentioning the fact that Elon Musk asked for it. Not even a passing mention of this.

Just, I believe satellite internet for rural Americans would be great. It's like, what? You're not talking about HughesNet, dog. Let's get real. You're talking about your buddy, Dr. Doge. Not even a mention. It's buried in the open meeting notes that they are responding to a petition by SpaceX. By saying, yes, of course, absolutely. Sounds great. Let's do it. Yeah. Yeah. Cool. Cool.

It's a good response. You know, he's a huge proponent of SpaceX over fiber. There's a lot of complaints about the Biden era bead program. Our buddy Ezra Klein is like basically selling his book by being like this program sucked. But the reason the bead program sucked is they took it away from the FCC. I'm

I'm not saying it didn't suck. I'm saying that one of the reasons it sucked is because the FCC was so politicized and stupid that it didn't have accurate broadband maps deliberately so that it could roll out the infrastructure that we were going to pay for. So they moved it to another agency that had to start the maps over without like Verizon and Comcast getting in the way, which is a real thing that like a GPI and Brendan Carr like advocated for back in the day that we're not going to, we're not going to demand these accurate maps. So Brendan is like,

satellites will fix it. Don't worry about this dumb bead program that didn't give you fiber. And you're like, dude, you don't even know where you're going. But anyway, he's going to, he's going to reform the spectrum and look, maybe, maybe this is the right choice. Maybe it is. But the reason that I'm calling it out is when you start with, I'm wearing the gold head of Donald Trump on my lapel and you end with, I'm doing what Elon Musk wants. You've lost your legitimacy as a regulator. Yeah. You just look corrupt.

And that is because you're a dummy, Brendan. So again, you know, Anna's welcome. Brendan, if you want to come on the show and explain your behavior or even just prove that you can complete a sentence without sucking up to Donald Trump or Elon Musk, you're welcome. I know you listen. I know you listen. Come on in, Brendan. It'll be fine. Water's warm.

That's been Brandon Carzadoni.

Wouldn't it be awesome if we stopped robocalls? And that hasn't worked at all. But then Brendan in here is like, we're giving Elon what he wants. And then he's just like, of course, spectrum is not the only area where we're moving fast. We're continuing the FCC's longstanding efforts to crack down on illegal robocalls. That segue doesn't make any sense. It's just him being like, but you hate robocalls, right? I'm going to fix them. By the way, it's 37 gigahertz spectrum, not 36. I apologize.

Brendan, if you want to, Brendan's like ready to tweet how wrong I am. Man loves a tweet. It is, by the way, true. Ajit Pai was all about stopping robocalls. It wasn't successful because he didn't actually regulate anyone. He just incentivized the industry to regulate itself. Straight up, not an exaggeration. I have gotten two robocalls since we started recording the podcast. It's very good. Anyway, that is Brendan Carr's Dummy, America's Favorite Podcast within a podcast. If you know where he got the pin from, let me know.

I'm sort of dying to. All right, David, we need some. I will make me like wear the pin on the show. If you get us one, I will make me like, it's the closest we're going to get to Trump or Brendan on the show. Let's be honest. All right. We need a palate cleanser. My friend theme music, TK, by the way. All right, let's, can we talk about tick tock for like 35 seconds? That's all I want. That's all I want to do. So the, the opposite of a pound cleanser. All right, well then we're going to get to the palate cleanser, but we got to talk about tick tock just for 35 seconds because we, we spent a little bit of time on it last week. Uh, nothing happened. Uh,

Everybody says it's against the law. Nothing is happening. Is that a fair summation of what's going on? Trump delayed it another 75 days. Pam Bondi, the attorney general, is out there essentially demanding that Apple and Google keep the apps alive, which is what they've been doing all along, saying we're not going to enforce this. You have to keep it in the store, on and on and on. And then Senate Democrats are saying it's against the law to do this delay, which, of course, is not going to happen.

is true. All you do is read the law and you know that that's the case. But I don't imagine any of that changing much of anything. It's a particularly bad situation for Apple and Google because they just get exposed to another 75 days of liability, just hoping that everything works out for them. They're just these like weird pawns trying to hope nobody notices them. And they're just gonna have to keep waiting this out because it's misguided.

Maybe never going to end. But they want tariff relief, too. So they've got to keep it like the level of just like base corruption here is wild, right? Like everyone has to fly to the golf course and be like, please, sir, not 104 percent. And they're all going to leave with like different numbers and compare them in the parking lot. They cannot take TikTok down because they want the number to be lower than the next guy.

It's this TikTok is doomed. That's my prediction. It is the ultimate pawn in this entire situation. China is not going to want to sell it. And at the end of the day, what they're going to say is turn it off. Like we don't want 104%. Donald Trump is not going to keep this promise. We're using China is going to say turn it off. It's the last bit of leverage. They did it once, right? They wanted to leverage the first time they just turned it off. The last bit of leverage you have, we don't want 104%. We want zero. TikTok's gone.

Trump will blink. That will work. I think you're right that that is that is like if you start at we're not going to release your movies here anymore. We're like a few escalations up to we're going to turn off TikTok and that will work. Yeah. In every way that China would need it to, it would work. Yeah. And China is not afraid to play these games. China is like, oh, you're the CEO of Alibaba. What if you weren't around for a while?

It is an authoritarian government. It's just a move they play. So we'll see. All right. Now can we have a palate cleanser? Yes, palate cleanser. I have two bits of exciting news for you before we leave. One is that it appears to be plausible, possible that Instagram is actually going to get an iPad app.

750 years after the launch of Instagram. Wait, can I tell you the funniest thing about this story? Yeah. It made me laugh so hard. This is also from the information, which is doing great work. It's in the context of a story that's about Adam Seri and Instagram being ready to be super aggressive if TikTok goes away. And there's a long list of stuff they might do, and they're like, it's time to ship that iPad app. And it's like, no, that's not it, guys. Right.

That's not... If TikTok goes away and your answer is, also, there's an iPad app now, I don't know if that's going to be it. But I do love it. I don't think that is how you do it. Like, TikTok famously also not a great iPad experience. But I think there's a case to be made for Instagram just attempting to be everywhere as it goes through all of this, right? And one way to...

make people happy and get a bunch of headlines is to launch an iPad app. Like, I think it's not crazy that Meta has been sitting on a finished iPad app for a long time, just sort of waiting for the right moment. I mean, there's no way this requires, I'm going to say like any work on their part, right? Like you like click a couple buttons that goes, it's wider.

You've got an iPad app. They were just like mad at Apple, right? Yeah, they just didn't want to do it. Yeah. Yeah. And I also, I was just talking to our friend Casey Johnston for a Verge casting that's going to run in the next couple of weeks. And one of her theories is that

Meta doesn't like iPad apps because they want you to be on your phone because the phone is the more addictive and harder to put down device, which just strikes me as a good theory. I have no evidence for that theory, but it's a theory that makes sense. But I think there are a lot of people who have iPads and want to look at Instagram on their iPads. And I think if you claim to be a media first thing that is great on screens, iPads kind of make a lot of sense.

Seems like a good idea. This also, I'm curious if this extends to threads, which also does not have an iPad app, which feels like a little embarrassing as well, particularly given that it is in the heat of battle with these other platforms. Instagram is at least like singular in its way. There's also no WhatsApp for iPad. Like Meta has made a calculated effort to not do the iPad. And I think...

It is in part because of this ongoing beef with apple, but I think there may be a moment here where it is so useful to do that, that they'll do it.

I mean, that's wild. Again, I just think it's very funny to like list of things to do if TikTok goes down and they're like, they move the card on the Kanban board from jail to immediate, you know? All right. What's the next one? The last one is the Pixel 9a, which Allison Johnson just reviewed for us. And it is great.

It is the mid-range phone you would want it to be. It comes in a cool color. It's like pink that I'm very into that Allison had. And this is like Google continues to make good phones. This phone is still $499. There's been some sketchiness with the release date and all this stuff, but the price is right at a time when it might not be for long. This is just a really good phone. And this has become the phone I tell most people who are not tech people, but...

but like Android phones to buy. And I feel like this easily continues that in a way that I find very exciting. Yeah, Google seems to be killing it with this series. And I think it's particularly interesting this coming like, what, a month after the iPhone 16E, which is $100 more expensive. And I think that Delta and just a couple extra little like...

You know, friendly tweaks on the 9A, right? The fact that it will do both fingerprint and face authentication, the fact that it has a higher refresh rate screen, go a long enough way to make it feel like a deal in the way that the 16E doesn't quite get there.

Yeah, I think Google, ironically, like the big tradeoff here that Google made is it does less AI stuff, which is a very surprising tradeoff for Google to make, but also a really good one. It has all the other stuff you'd want your phone to have. It just doesn't do quite as much of the AI stuff. And spoiler alert, most people are pretty willing to make that trade.

It turns out. All right. I just want to end the episode by pointing out that in the time that we've been talking, the tariff on China has risen to 145%. 145? The White House has clarified that the tariff on China is now 145. You know what's manly, Eli? 1,000%. Real men do 1,000%.

I'm just saying it's a very loud V8 engine, my friends. Wait, can I ask one more question about phones before we go? This is the thing I've been thinking about. I'm curious what you guys think. I think my sort of ongoing thesis for a while now has been that phones are not actually that price sensitive anymore. That, you know, you get the trade-ins, you get the financing that like actually the sort of $100 delta between one phone and another is not that meaningful anymore.

But I wonder, in a tariffs-y world where people are more price conscious about everything, are we as people who write about these things going to have to care about price in general more than we have in the past? It's interesting because we run a reviews program, and so much of a reviews program is like, is it worth the money? And so I want to say the answer is yes, because that just changes the heart of the thing that we do. But I also think what's going to happen is like,

There's going to be more bloatware on the phones, right? Like people are price sensitive, especially right now. And I, you're just, I think you're just going to be like, you open your pixel line a and Google is like, do you want three free months of max? Yeah. Right. Like it's just that thing is going to start happening in extremely weird ways or they're going to lower the base storage and be like, do you want Google one? Like, you know, like there's something shittier that's coming that like, we'll keep the price the same.

I'm not sure what it will be, but it feels like they're not going to up the price, right? Because that's a drastic measure. Yeah. And it is easier to sneakily make the phone worse than to sneakily make it more expensive. That's definitely true. Yeah, like if you're any one of these big – if you're Apple or Google, you can go to one of the big carriers and be like, here's what we need to do. We need to adjust the terms of our revenue sharing.

And so the phone contract lengths will be six months longer or add on a provider service upcharge fee that's labeled like the cook doctrine. But it doesn't matter. You know, like there's just other places to sneak the money in that isn't the selling price because of the way we've structured all of these things. I think I buy that. It feels like it feels like bad news. Yeah. But I think that I think that is where this goes. Like, we're going to see.

Chris Mims, my old colleague at the Wall Street Journal, was tweeting about the thing people don't realize is not that a lot of stuff is going to get more expensive. It's that a lot of stuff is going to go away. And I think that and the sort of inshitification of things ahead of their time are two sort of non-obvious things to look out for here. Yeah. By the way, the 145, they clarified it's because of fentanyl, which is one of those things. It's not a crisis coming from China. Yeah. What's the trade deficit on fentanyl?

It's very confusing. Like, there's a hole in the heart of this whole argument that's like, oh, your emergency is fake. So we'll see. By the time you listen to this, the number could have gone up or down. Who knows? Email us and tell us what the number is right now as you're hearing this. I desperately want to know. Yeah. All right. We got to get out of here. That is the VergeCast for today.

And that's it for The Verge Cast this week. And hey, we'd love to hear from you. Give us a call at 866-VERGE-11. The Verge Cast is a production of The Verge and the Vox Media Podcast Network. Our show is produced by Will Poore, Eric Gomez, and Brandon Kiefer. And that's it. We'll see you next week. ♪