Iran Deadline Looms for an Economy on the Edge

Ed Elson and Justin Wolfers break down the state of the U. S. economy, including the latest jobs report and how the Iran war is shaping the economic outlook. Then, Paul Kedrosky joins the show to unpack OpenAI and Anthropic’s financials, and where their AI IPO race stands. Finally, Ed shares what OpenAI’s acquisition of TBPN reveals about the future of media. Justin Wolfers is a professor of public policy and economics at the University of Michigan. Paul Kedrosky is a Managing Partner at SK Ventures. Subscribe to the Prof G Markets Youtube Channel Check out our latest Prof G Markets newsletter Follow Prof G Markets on Instagram Follow Ed on Instagram, X and Substack Follow Scott on Instagram Send us your questions or comments by emailing [email protected] Learn more about your ad choices. Visit podcastchoices.com/adchoices

Speaker names marked ? are inferred from speaking time and the show’s known hosts — they are not voice-verified.

• 0:01Speaker 00

  This episode is brought to you by Michelob Ultra. Good times mean competitions between you and your friends. Like seeing who can throw some pigskin the farthest or rush the field the fastest. Winner gets an ultra because it's not just a superior light beer. It's also giving fans superior access to their favorite sporting events. Superior is worth playing for. Visit michelobultra.com/superioraccess to enter. Enjoy responsibly. Copyright 2026 Anheuser Busch InBev Michelob Ultra Beer, St. Louis, Missouri.

• 0:30Speaker 01

  Flowing ad budget on metrics that look great till the CFO sees them. That's bull spend and marketers are calling it out in dashboard confessions.

• 0:39Speaker 02

  I remember telling my boss, it'll be good for the brand when leads were slow. Yeah, it it wasn't.

• 0:47Speaker 01

  Cut the bull spend. LinkedIn lets you target by company, job title and more. Advertise on LinkedIn. Spend $250 on your first campaign and get a $250 credit. Go to LinkedIn.com/campaign. Terms and conditions apply.

• 1:01Speaker 03

  Low energy, unfocused, foggy. You might be dehydrated. Gatorade Zero powders have been scientifically designed to help improve hydration balance in the body with a clinically proven electrolyte blend and no artificial colors or flavors. Great tasting hydration, no matter where you're going. Shop now online or at retailers nationwide. Gatorade hydrates better than water. Is it in you? Symptoms noted are signs of mild to moderate dehydration. Consult a healthcare professional if symptoms persist.

• 1:29Ed Elson?

  Today's number 60. That's how many days it's been since the Dow hit 50,000. The index has since fallen to roughly 46 and a half thousand, which means we were only three and a half thousand points away from keeping our attorney general.

• 1:49Speaker 05

  If money is evil, then that building is hell.

• 1:57Ed Elson?

  Welcome to Profty Markets. I'm Ed Elson. It is April 7th. Let's check in on yesterday's market vitals.

• 2:05Ed Elson?

  The major indices rode out a volatile trading session to end in the green. Traders wavered between hopes for a ceasefire and fears of escalation from President Trump. Brent crude climbed again above $110 per barrel. Meanwhile, Treasury yields and the dollar were flat.

• 2:23Ed Elson?

  Okay, what else is happening? Trump's deadline for Iran to open the Strait of Hormuz is tonight at 8:00 PM Eastern. So far, both sides have rejected each other's proposals to end the war. President Trump escalated his threats against the country on Sunday and yesterday, he warned that the entire country could be, quote, "taken out" as early as Tuesday. The conflict has already triggered the largest oil supply disruption in the history of the global market, according to the IEA. That inflationary pressure gets its first real test on Friday with the release of the March Consumer Price Index data. We also saw the March jobs report last Friday, which provided a snapshot of where the economy was before the war.

• 3:04Ed Elson?

  The economy added 178,000 jobs and the unemployment rate ticked down to 4.3%. However, the hiring rate fell, the participation rate declined, and wage growth slowed. Meanwhile, the February numbers were revised from a loss of 92,000 jobs to a loss of 133,000 jobs. A lot there. Here to get the state of the economy, we're doing an economic update with Justin Wolfers, professor of public policy and economics at the University of Michigan. Justin, great to have you on the show again. We wanted to get your update on the economy now that we're in week six of this war. They said it was going to be four weeks or four to five weeks. We are now extending beyond that. What do you make of what's happening right now?

• 3:49Justin Wolfers?

  Ed, I think the good news is you and I have had a rehearsal for this. So you, you know, we spent all of 2025 saying the president is on edge, the president might be about to do dramatic things that could upturn how we think about global commerce. The jobs reports looked weak and then they look not quite so weak and we've been on edge. And one thing I can say, Ed, I'm a little old at news, so grasshopper, it's not always this way. There was a time when I'd be able to do a long and serious interview with it about the economy, understanding that it would have a shelf life of more than six hours. That's not the world we're in right now. So let me go back and answer your question.

• 4:35Justin Wolfers?

  I think the the jobs report was unabashedly good news. If you didn't update after that and say the economy is a little more robust than I had feared, then you're not paying attention. Equally, when you say that we created, was it a hundred and seventy eight thousand jobs in a month?

• 4:53Ed Elson?

  Yeah.

• 4:54Justin Wolfers?

  In the current moment, if that were the true underlying pace of job growth, you would see me gritting from ear to ear. I'm not gritting from ear to ear. Um, there's some underlying weakness there, which is some amount of this was, you know, a bunch of folks are on strike, one month aren't on strike the next. That gives you a lot of the healthcare. A bunch of people are out because of bad weather. Um, and they're now back because of good weather. And that also plausibly explains quite a lot of what's going on. So it's sort of a fantastic number with a weak underbelly. Um, but does suggest that at the very least, the economy is running somewhat out. You know, it doesn't look like we're in a recession. Um, our expectations for the rate of employment growth should be substantially lower

• 5:39Justin Wolfers?

  because we have no population growth. Um, so the US economy is to some degree the little engine that could. The thing I like to see most is that the unemployment rate, um, I think we should all keep our eye on the unemployment rate, especially right now, because looking at employment only matters is only useful if you know how many jobs we ought to be creating.

• 6:00Ed Elson?

  Right.

• 6:00Justin Wolfers?

  And because of the shock to immigration, how many jobs we need to create is much, much less clear. So the unemployment rate sort of tells us how we're doing relative to how we got to do. And it's moving roughly sideways. Um, that's certainly not a failing grade. Um, it's not, it's not incredible success, but you know, we're doing okay coming. So what that says is on the eve of the Iran war, the US economy is doing relatively well, better than pessimists, and I've been a pessimist over recent months. Would have expected, um, and not as well as the optimists of Boston.

• 6:40Ed Elson?

  Yeah. A, a couple of questions that I, that come to mind. The first is, and, and, well, I'll start with the first question.

• 6:51Justin Wolfers?

  It's a good place.

• 6:53Ed Elson?

  The revision, we keep on getting these down revisions. And so I look at the numbers that are coming out from this employment report. Hundred seventy eight thousand jobs added. And it seems like the way this always goes is we all go, oh, it's not, it's not so bad. It's actually pretty good. And then we come back to it a month later and we learn actually it's like forty to fifty to sixty thousand jobs lower than we originally projected. And in fact, Jerome Powell told us as much. He was literally telling us in the previous Fed meeting that we're basically overshooting the jobs added by around, I think he said sixty thousand jobs each time. So I guess my first thing, I see the number and I'm like, well, I don't even know if

• 7:35Ed Elson?

  this even matters or if this even means anything, because I'm pretty sure that it's going to get revised in the next, in the next round. That's the first thing. And then the second thing is, well, now this, we also have this war with Iran, which is kind of a big deal.

• 7:48Justin Wolfers?

  We have to stop at one thing.

• 7:50Ed Elson?

  Okay, let's go, let's go with the first thing.

• 7:53Justin Wolfers?

  Look, so what I don't want you to do is see revisions and say, oh, none of this matters. Because what you're looking for is the least bad indicator in a universe of terrible indicators. If it helps you see it that way, that's fine by me.

• 8:10Justin Wolfers?

  It's hard to see through fog. It's hard to measure an economy of the size and scope of the United States. It's even harder when they're in the middle of so much change, which, and I think this is where you're going with your second question, what happens in March? You know, what happens before the war may be much less relevant than what happens after the war. We could say the same thing with the trade war as well. So it's hard. So you're sailing through fog. What do you do? You pull out your flashlight. The thing you discover about your flashlight is it helps you see about 12 feet in advance. And a lot of really interesting stuff is like 13 or 14 feet in the future. And that's sort of your point about the revisions, which is we have a rough sense that there's not a massive iceberg right there.

• 8:53Justin Wolfers?

  Do we know that we're not going to run into a couple of penguins on the way? We could. Yeah. They might be there. Um, so even with all the revisions, what's true for any sort of serious analyst of business cycles is if you look at the ups and downs of every other economic indicator, payroll's still, even with its revisions, is timely. Yep. Right, just came out. Uh, extremely reliable, even though you just described as extremely unreliable, that's correct. There's some distance between it and the truth. But that distance is shorter, smaller than it is for anything else. And with this extraordinary track record over time. And it's the most watched market indicator, not because guys in markets like to do what everyone else is doing,

• 9:38Justin Wolfers?

  but because it still has more signal than anything else. Um, so, um, and then I just want to separate one other issue you described. So there's the revisions and the revisions are very large. They're actually separate from what Jay Powell was talking about. Jay Powell, where he said payroll's is being overestimated, he's talking about benchmark revisions. So let's nerd out, I know you love to nerd out, three ways in which each month we learn more about the past and we go back and revise it. The first is what we think of as being revisions. We use the word revisions for it, which is there's a bunch of firms that didn't send in their form last month, telling us how many people were on their payroll. They eventually get round to it and we figure rather than using statistical extrapolation, let's count what they actually told us.

• 10:24Justin Wolfers?

  And so then we have to update what we thought happened last month. That's a standard revision, right? Second thing, not many people appreciate the following, is also seasonal adjustment, which is a very complicated beast. We update the seasonal adjustment factors every month.

• 10:40Justin Wolfers?

  And so some of the, there's a further revision. There's a revision to the non-seasonally adjusted data, that's the new forms that come in. There's also a revision, usually pretty slight, to the seasonal factors. So how much of it we think is the ups and downs of summer and winter? And then the third thing that happens is what we have at any point in time is a sample of the population of all firms sending in their forms. But to know what that means for the economy, we have to know whether that's a one percent sample or a two percent sample. We have to know what share of the economy is being featured. And we also have to make sure that it's representative of small firms and large firms and all of that stuff. That's what's called the benchmark revision. The benchmark revision is we go to a near universe of firms and we say, oh, this sample

• 11:25Justin Wolfers?

  we have, how representative is it? Let's rescale things once we know how representative or unrepresentative it is or isn't. The near universe is called the Q C E W. It comes from unemployment insurance records. And what we saw through twenty twenty four and early twenty twenty five was that there were fewer firms than we'd implicitly been assuming. And we will get a benchmark revision much later in the year or some distance away from it. But we see early data from the numbers that are used as an input to that benchmark revision. And that's what Jay Powell's talking about, where he says, I'm expecting another revision to come. He's not talking about the monthly revisions.

• 12:09Justin Wolfers?

  He's talking about the annual benchmark. It all matters because it's all, he's saying, what's going on right now probably overstates the extent of job growth.

• 12:18Ed Elson?

  Right.

• 12:19Justin Wolfers?

  Um, but it's not going to get washed out next month. It'll get washed out. It'll get fixed up in several months with the benchmark revision. Sorry if that was boring.

• 12:27Ed Elson?

  No, no, it's, it's helpful to understand exactly what goes into this stuff. And it's interesting because these things, they're so important to the cultural conversation and the economic conversation. Then there's a lot of caveats and details that sort of get lost in the complexity of it all, but people should really know about it. I do want to pivot to what's happened with Iran here because what is so interesting is that employment report comes out, big deal, very important for someone like you as an economist. But then. We also see this message, I'm going to read it verbatim from the president, quote, "Tuesday will be power plant day and bridge day, all wrapped up in one in Iran. There will be nothing like it. Open the fucking straight, you crazy bastards, or you'll be living in

• 13:12Ed Elson?

  hell. Just watch. Praise be to Allah." President Donald J. Trump. I guess my question for you, which one is more important as an economic piece of data or economic piece of news to you as an economist, that tweet or the employment report?

• 13:29Justin Wolfers?

  The tweet. The tweet for two reasons. One, this tells us something about the future of the war. The war is overwhelmingly the most important economic thing going on right now. You don't have to be too much of a careful student of history to understand that wars have enormous implications for politics, for economics, for geopolitics, for our ability to do business with each other, for our future defence budgets, for those of our allies and our former allies, for our fiscal situation, and in fact, for our monetary situation. So the way this war will have direct effects through oil and indirect effects through geopolitics,

• 14:12Justin Wolfers?

  defence and all the stuff that I feel somewhat unqualified to opine about, the problem, of course, is the folks who understand that stuff don't understand economics very well. No one's really qualified.

• 14:22Ed Elson?

  We need someone to talk about it. I think you're the guy. You're as good as we're going to get here.

• 14:27Justin Wolfers?

  Well, I'll be honest about what I don't understand. I think that's probably the most useful role that I can play. But I think there's a second thing, no less important that we learned in that tweet. And we saw this, the same dynamic actually play out during the trade war. During the trade war, there were some really important moments, particularly around liberation day and the aftermath, where the market movements were enormous, even though what the president was doing was pretty small. And I think what was happening was markets were updating on this fundamental idea of how competent is this administration? How much do I trust it to react in a way that defends America's interests to challenges that come our way? And some of the lunacy around liberation

• 15:08Justin Wolfers?

  day led to the Sell America trade, if you remember that, that language. And it led to, I think, an outright lack of trust in the administration.

• 15:19Justin Wolfers?

  When you have a president using, you know, I'm a parent, Ed, my kids would be in trouble for that tweet. Um, they'd be in trouble for swearing, but fuck that, I'm okay with swearing. Trying to offend people about their deeply held religious beliefs at a moment where what's actually more important is that you both find a path forward towards peace is inane, insane, and does not serve America's interests at all. And so what we're learning from that tweet is not just something about the process of this war, but the what's going to happen over the four years, right? Um, how much better do you feel about, I don't know the next thing to come around the corner.

• 16:04Justin Wolfers?

  What if it's a drought? What if it's an earthquake? What if it's a pandemic? What if it's, uh, I mean, just a productivity slowdown? What if AI does something crazy, whatever it is? How reassured do I feel that the president is going to make smart choices with our enormous military and scientific might in a way that leads to a more productive future? I feel much less confident. Um, there are very real questions about the president's judgment. There are questions about his age, there are questions about his mental acuity, and if not that, then there are questions about his political strategies and his understandings of economics.

• 16:47Justin Wolfers?

  So I, I made that list broad so I can include folks who are somewhat on the conspiracy side, but also at the other end, just these are questions of judgment. Do you feel well led? Do you feel confident? Do you, uh, think now is a good time to invest in the United States? Or would you like to see some of these big issues resolved, um, before you're going to lay down millions or billions of dollars on a major investment here?

• 17:14Ed Elson?

  Yeah. Just before we let you go, I mean, the, the idea is that by Tuesday evening, we have some sort of a deal and some sort of a ceasefire in some capacity, this comes to an end or we get a signal that it's coming to an end, whatever. That's the idea. I know you're not a geopolitical expert, I know you're not a military strategist, but you are a smart person who pays attention to what's happening and pays attention to the implications of what will happen.

• 17:42Ed Elson?

  Do you think that this is going to end anytime soon?

• 17:45Justin Wolfers?

  Um, I hope and pray, and I'm not even religious. I think the stakes, you know, I think the only smart thing I can say that's moderately informed, that comes from my expertise is the stakes here are really, really large. I will spend the next 24 hours on the edge of my seat. I am deeply worried. I worry partly because I care about the consequences for the American economy. I worry partly because I care about the consequences for the global economy. And remember that many of the consequences of our actions are being felt two times, four times, 10 times larger than other countries.

• 18:23Ed Elson?

  Yeah.

• 18:24Justin Wolfers?

  And I worry because there are people around the world right now living in fear. And we should never lose sight of that. And great leadership is about removing that fear, giving us all the capacity for joy and safety and freedom. And it's a really, really critical few moments right now for trying to help spread that freedom around the world.

• 18:52Ed Elson?

  Justin Wolfers, professor of public policy and economics at the University of Michigan. Justin, always appreciate your time. Thank you.

• 19:00Justin Wolfers?

  Pleasure, mate.

• 19:02Ed Elson?

  After the break, an update on the AI IPO race. And if you're enjoying the show, please follow our new Prof G Markets YouTube channel. The link is in the description.

• 19:49Speaker 07

  The new LinkedIn Hiring Pro can't undo your last hire. The human postponer. They were the master of one phrase. I'll circle back on that. But three months later, you were the one doing all their work and wondering how big that circle is. But LinkedIn Hiring Pro can take the hiring load off your plate by automating the hiring busywork from the initial job post to scheduling interviews. Hire right the first time with LinkedIn Hiring Pro. Post a free job today at LinkedIn.com/quality.

• 20:20Speaker 08

  Low energy, unfocused, foggy? You might be dehydrated. Whether it's hot yoga, over 100-degree weather, or too much sun. Gator Light, with a specialized blend of five electrolytes, was scientifically designed to help move fluids into the body faster for rapid rehydration. Shop now at retailers nationwide. Gator Light hydrates faster than water. Is it in you? Symptoms noted are signs of mild to moderate dehydration. Consult a healthcare professional if symptoms persist.

• 20:53Ed Elson?

  We're back with Prof you markets. OpenAI and Anthropic are racing towards what could be the largest IPOs in history. The Wall Street Journal obtained confidential financial documents from both companies, giving us the first real side-by-side look at their books. Both companies are growing quickly. OpenAI has reached $25 billion in annual recurring revenue, while Anthropic is at $19 billion. But they both share the same problem. The cost of training these new AI models is devouring their margins. OpenAI expects to spend $121 billion on computing power in 2028. Anthropic plans to spend roughly $30 billion. Meanwhile, there is fresh drama inside OpenAI.

• 21:36Ed Elson?

  The company's CFO, Sarah Frye, reportedly told colleagues that the company isn't ready to go public this year, and that put her directly at odds with Sam Altman, who is pushing for a listing in Q4. Since those comments, he has reportedly cut her out of key investor meetings. Lots to discuss here. Here to discuss the state of the AI IPO race, and he's already laughing, is Paul Kudrowski, managing partner at SK Ventures. Paul, thank you for joining us. I guess we'll just start with what's making you laugh right now.

• 22:09Speaker 09

  You know what it reminds me of is there's this great horror movie trope where the, you know, the heroin is inside the house towards the end of the movie and she's talking about how like what's going on and she feels really threatened and somebody jumps on the call and says, "The call's coming from inside the house. Get the hell out." This is kind of that, right? I mean, it's literally, I've been around this stuff for decades, to my shame, and a public, well, a public IPO, an IPO candidate's company's CFO, saying that we think we should probably not go public, just doesn't happen, right? The job of the CFO is to say, "Yes," and alternatively, "We could be sued for that." And that's really it. That's all you're supposed to say as a CFO. You really, you've got like

• 22:53Speaker 09

  sort of two options, right? Option three is not, you know, I kind of feel a little bit meh about the IPO. I think that option three doesn't really exist. So this is really unusual. And to be somewhat serious, I mean, the idea that someone would say this at this point in the process and then potentially be cut out of conversations with respect to the book building and putting together documentation with respect to the IPO, and that someone happens to be your CFO, which is the primary reporting officer from an SEC standpoint in terms of bringing the company public, is, to use the technical term, ridiculous. I mean, it's just ridiculous. So.

• 23:28Ed Elson?

  Well, I mean, just so that people know, I mean, you were an engineer, you were a tech equity analyst, now you're an investor. So that you have seen this world before. I guess the question becomes, if it is ridiculous, and I happen to agree.

• 23:43Speaker 09

  Yeah, hard not to.

• 23:45Ed Elson?

  Is it because the CFO is the problem here? Is it because she's doing her job wrong? Or is it because the company is the problem? The company isn't ready to go public. It's in such a bad position that she is actually maybe doing a good thing by saying, "Hey, we really shouldn't do this."

• 24:01Speaker 09

  She's right, but it doesn't matter, right? If the CEO wants to take the company public, and your view as the CFO is, "We should not go public," then you're at the wrong company. I mean, it's literally that straightforward, right? That you can't have that kind of disconnect. This isn't, you know, providing guidance to Pete Hegseth before he sends troops into Iran and saying, "You know, Pete, I'd rather we didn't do this right now." It's like, "No, dude, you're fired."

• 24:22Ed Elson?

  Yeah.

• 24:23Speaker 09

  I'm not, that's not the advice I want right now. Here's the advice I want right now. And the same thing applies in the context of OpenAI. You're simply telling people in a very strange way that you're at the wrong company. Which, by the way, to your point, to your point, though, she's not wrong. I mean, there are all kinds of things that were just, you know, wacky in the Wall Street Journal article, the accounting treatment of frontier training, the provisions with respect to suspending the NASDAQ requirement to becoming a part of the index. All of these things raise all kinds of problems that will all become obvious over the next 12 months. So, you know, the idea that in some sense this company should not be going public right

• 25:04Speaker 09

  now is much broader than just, you know, their unit economics suck or some such thing like this. It's a much broader problem in that. It's more systemic about the nature of what's going to happen when multiple trillion-dollar market capitalization companies come to market in the course of 2026. This is like an economic shock treatment because you have to think about where's the money going to come from that flows into these? Well, it's going to come out, come from a host of other companies who happen to be very liquid, traded and overlap. Well, who are those guys? It's Google, Microsoft, Amazon and others. So money's going to flow from those places into these places because in a sense you're being forced to by putting them into the index so quickly because index ETFs must immediately purchase the stock. They don't hold cash.

• 25:49Speaker 09

  They're not like somebody with like a bag of money in the basement. They have to sell something else to move it over there. And so you're going to see this giant series of tsunamis of cash flows that are going to be wildly disruptive all because of this thing, this behemoth. These three behemits are coming public this year.

• 26:04Ed Elson?

  So much there and it's so interesting. Almost your answer to my question is like it's both a problem that the CFO is making a mistake and also the company is not in the right position. It's sort of like a double whammy of badness.

• 26:17Speaker 09

  Well, in a triple whammy, it's also systemically problematic because of the amount of money is going to cause, you can model it out and hedge funds are doing it right now. I mean, one of my side gigs is talking to a lot of some of the largest hedge funds on earth and they're already looking at this and saying, okay, the outflows from these places are going to do what to their share prices? Okay, then that's, that has to mathematically happen because I have to fund my purchase because of this new requirement that they can be in the index very, very quickly. So what does that mean? It's kind of like a giant index rebalancing exercise. It's going to have consequences for one group of companies and consequences for another group as this this great big pile of money swishing, you know, goes swooshing through the economy. So there's there's the high level issues and lower level issues, not least of which is,

• 26:58Speaker 09

  you know, this bizarro method that they want to use to handle to handle frontier model expenses. So there's, you know, it's just the issues are legion.

• 27:06Ed Elson?

  Tell us a little bit about what you make of the business itself. I mean, they just crossed $25 billion in ARR. They are the number one AI company. They have a ton of users. I guess it starts to get a little crazy when you look at the valuation, $852 billion as of this most recent round. I mean, what do you make of the business?

• 27:25Speaker 09

  Revenue, yeah.

• 27:26Ed Elson?

  Yeah, what do you make of, I guess, the valuation and also the fundamental business itself?

• 27:32Speaker 09

  It's the unit economics are negative, meaning they, you know, they lose money. They lose money on every unit they sell. That hasn't changed. And the way they try to get around that is by excluding their most, their largest, most material and most predictable cost, which is frontier model training. So the line of patter is that, and this was part of the Wall Street Journal piece, was let's look at their earnings excluding frontier model training, excluding model training. And on that basis, I think it was OpenAI said in the piece, or the implication was in the piece that they'd be cash, could be cash flow positive in a very modest way. I think it was in 2029 or something like that, if you exclude those costs. But that's a ridiculous exercise because those are not costs you can exclude because that is a persistent annual expense that they concede in the piece is actually both

• 28:16Speaker 09

  accelerating and becoming more expensive. And so that's not a cost that's going away. So excluding it is a fundamental misunderstanding of the nature of their business. So to do that, you can, or by doing that, you can make the unit economics appealing, but it's a misnomer because the unit economics are driven by these, by the models themselves, by their own admission, that if they were to stop spending on models today, subscribers would move somewhere else.

• 28:41Ed Elson?

  Yeah, you're reminding me of just sort of a very simple truth, which I feel like was top of mind for me and for many others, maybe a few months ago, which is like this business actually isn't a good business when you just. Look at how it works. As you say, look at the unit economics. And in a way, I've sort of forgotten that over the past

• 29:00Speaker 09

  Yeah.

• 29:01Ed Elson?

  couple of months, I think, because

• 29:03Speaker 09

  Right.

• 29:04Ed Elson?

  there's a lot of other stuff that's happening. Also, I mean, Anthropic has made a pretty incredible push. And we've seen a huge increase in their error recently, especially after what they had with the Pentagon. I mean, I tend to sort of draw a line in the sand between OpenAI and everything else. But how do you view the AI ecosystem? Are you as bearish on Anthropic, for example, as you seem to be on OpenAI?

• 29:35Speaker 09

  No, but only because OpenAI is has probably one of the more bizarre management structures out there. There's a new New Yorker piece out there talking about this as well. And I don't see how they're a long term survivor here for a bunch of dysfunctional reasons, not just the unit economics, but also the management structure. So, no, not as bearish, but they have the same unit economics problem, which is to say that they're faced with this giant problem of having to train very, very expensive models to more or less stand still. And there seems declining returns to training those models, which are largely masked by the emergence of these very effective so called orchestration layers like Claude Code or Codex, which in a sense are like, I often argue, are like really effective nannies who have to deal with really bratty kids. The bratty kids are the models.

• 30:20Speaker 09

  The really effective nannies are Claude Code and Codex. Most of the value is shifting, if you think about it in operating system terms, are shifting towards these orchestration layers. And I sometimes joke that the model company that survives will be the first one to abandon models and just sell orchestration layers and things on top, because the models are the problem. That's the capex. That's what's breaking the unit economics at Anthropic and Code and OpenAI. So in a weird, almost, you know, as a thought experiment, is at least we're thinking about, you know, what happens as this realization finally hits hard and these companies are public and they're finally being really and truly punished for this massive capex. I wouldn't be surprised to see them rapidly shift and just become rappers on other things.

• 30:59Ed Elson?

  What do you, how do you think this ultimately all plays out? Because essentially what you're describing is yes, this, this technology is quite incredible. And yes, a lot of people are using it.

• 31:07Speaker 09

  Yeah.

• 31:07Ed Elson?

  But from a business perspective, it doesn't make any sense because it's too expensive to operate and to train and to run. And therefore, if we were to play this out over the course of several years, my mind goes to maybe AI just doesn't really happen in the way that we expected because maybe in some sense, the money runs out. I mean, the VCs and the Nvidia's of the world are essentially paying for this right now.

• 31:34Speaker 09

  Yeah.

• 31:35Ed Elson?

  But I guess what happens 10 years down the line?

• 31:37Speaker 09

  Yeah, well, it doesn't, it's not gonna take 10 years. We felt like we were so lucky as to have it play out for 10 years. For me, it ends up looking kind of like the enthusiasm that led to rural electrification in the US in the 1920s, where there was this incredible explosion of excitement. It was going to be transformative. It was going to be a growth industry. It was going to be ubiquitous and inside of everything. And it kind of disappeared, like you, you can't really see it anymore. Electric, electricity until at least the AI explosion of demand for training was kind of disappeared into the woodwork. And I think AI will rapidly do the same thing. It'll be almost like a utility like service. And it may be a utility like service that kills us all and turns us into paper clips, but it's still going to be utility like service. And so in that sense, the valuation accorded to it will change really rapidly.

• 32:19Speaker 09

  So my, my hunch is it's one of these classic industries that you see kind of a double and then, you know, then everything just sort of the wheels fall off and everything falls by half or more. And so I think there'll be this incredible retail driven enthusiasm for these IPOs. It'll cause a huge amount of money sloshing around in the markets, some huge dislocations in some of the other names out there. And then eventually these guys are getting punished for their capex.

• 32:39Ed Elson?

  Right.

• 32:40Speaker 09

  And what do they do about it? Do they cut it? Or do they, or do they do what I'm suggesting, which is that some of them will actually move to just being pure wrapper companies? I have, you know, that's, that's a viable option.

• 32:49Ed Elson?

  Right. What, just before we let you go here, how do you, something I often think about with OpenAI, with Anthropic, even with something like XAI, I'd be fascinated to know how these companies, how these, the shares would be performing if they were publicly traded. That's like my, always my number one question, whenever I see any news related to OpenAI. Valuation keeps going up and up and up, but it almost seems like a fake valuation.

• 33:13Speaker 09

  It's going to be fake valuation. And so you're seeing that already in secondary markets, though. In secondary markets, Anthropic is performing much better than OpenAI. Demand for OpenAI shares in the secondary market, which is to say, among people who already hold the shares through various other transactions, is already much weaker than the demand for shares of Anthropic. Now, some of that's purely valuation driven, but it's also a nice tell about the nature of how institutional investors and others already feel about the relative merits of the two companies.

• 33:39Ed Elson?

  Yeah. So I assume that, I mean, it sounds like you think the company is going to go public. A lot of retail demand because what other AI company are you going to invest in? Very exciting. But then.

• 33:50Speaker 09

  Well, because the Chinese AI companies are doing really well and people want, people want, there's a huge appetite for financial instruments. Just look at what happened with five or six now Chinese retail AI IPOs. The same thing is going to happen here.

• 34:02Ed Elson?

  Yeah, really, really interesting. Okay, Paul Kudrowski, managing partner at SK Ventures. Paul, that was very, very interesting and we'd love to have you back again soon.

• 34:12Speaker 09

  Sure.

• 34:18Ed Elson?

  In related news, OpenAI is getting into the podcast game. The AI company just acquired TBPN, a technology talk show that's become quite popular in Silicon Valley. How much did OpenAI pay? Well, we don't know as the sum was undisclosed. However, there have been some rumors, according to the Financial Times, the deal was in the, quote, low hundreds of millions of dollars. And I was also told separately by someone who knows people familiar with the deal that the company was bought for $200 million. So we don't know the exact numbers, but either way, OpenAI bought a podcast and they paid a lot of money for it. And it is especially a lot of money when you consider the fact that actually this podcast, while it is popular.

• 35:04Ed Elson?

  It isn't that popular. In fact, their live streams, which is how they distribute the program, only receive roughly 7,000 views per episode. For context, this show, Prof G Markets, reaches roughly a quarter of a million people per episode. So why would OpenAI pay this much? Why would they pay what Spotify paid for the Joe Rogan podcast, for example, which was the most popular show in the world at the time? Well, there are a few reasons. Number one, OpenAI wants influence in the tech community specifically, and that is a very hard demographic to reach, which TBPN really owns. Number two, they probably wanted some marketing savvy in their organizations, so they paid a premium

• 35:48Ed Elson?

  for that. And most tech commentators have talked about this. But the third reason is less discussed, and it is perhaps the most important reason. And that is that OpenAI actually wasn't paying for a podcast, per se. What they were paying for was clips. Yes, clips, those short, one-minute videos that you see on your social media feed every day. Those are actually very important, and they are, in fact, the backbone of TBPN, along with every new media organization on the rise. And here's the secret. While TBPN's live streams average only 7,000 views per episode, their clips average more than a

• 36:31Ed Elson?

  quarter of a million views. And in addition, unlike most media companies, this media company sells their ads directly into their clips. They'll say things like, this clip was brought to you by, insert, large corporation. And that is how they're making their money. That is why they just got bought for $200 million. It is an incredibly smart move. But it also says something about the media environment in which we now find ourselves. Think of any rising star in media right now, and ask yourself, what do they all have in common? How did, say, Nick Fuentes, the far-right white nationalist, become the viral sensation that he now is, the guy who is feared and talked about by political

• 37:16Ed Elson?

  leaders on the left and on the right? Did he do it with his live stream? No. His show only gets around 20,000 viewers per episode and it's only available on a platform called Rumble. No, he did it with clips. And that is every day, Nick Fuentes clips go viral on X and on Instagram and on TikTok. In fact, one of his most recent clips received nearly 11 million views in less than two weeks. That is equivalent to the number of people that Fox News reaches over the course of two months. The same goes for the latest Gen Z looks maxing star, Clovicula, who you might have heard of. His livestream averages only 16,000 viewers. That's really nothing, but his past 10 clips on TikTok have reached

• 38:01Ed Elson?

  70 million viewers. That is more than the population of France. Almost every new media sensation has leveraged this format, from Hassan Piker to Andrew Tate to Mr. Beast. In fact, it has now created an entirely new economy. Content creators are now paying agencies to create clips of their content and distribute them across the internet. Andrew Tate, for example, actually pays his community of fans, of which there are 100,000, to post clips of him. Mr. Beast just launched a clipping platform where he pays clippers to promote his new shows. In fact, one livestreamer called Neon actually revealed on his show how much he pays his clippers. Someone asked him, how much do you pay all

• 38:46Ed Elson?

  of your clippers each month? To which he replied,

• 38:48Speaker 10

  It was anywhere from 750 to like 900 a month.

• 38:53Speaker 11

  Why? What are they posting? I'll be your clipper. What the fuck?

• 38:56Speaker 10

  No, you're doing great.

• 38:57Speaker 11

  I will clip you.

• 38:58Speaker 10

  No.

• 38:59Speaker 11

  A million dollars?

• 39:00Speaker 10

  No, well, it's spread out with around like 400 people.

• 39:02Speaker 11

  Okay, so 400 people, what's like their base pay?

• 39:05Speaker 10

  Uh, it just depends on their views. So it's all based on how many views they're getting, but my top guy this month is getting like 170.

• 39:11Speaker 11

  170,000 a month?

• 39:12Speaker 10

  Yeah.

• 39:13Ed Elson?

  So there you have it. Content creators are now paying people nearly a million dollars a month just to create clips, which means that we have now officially entered into the clip economy. Clips are no longer a secondary medium. They're not promotional material for your main show. No, clips are your main show. They are where all the action is happening and increasingly where all the money is being made. So what does this mean for media companies, companies like us? Well, it means you need to get better at clipping. And more importantly, you need to get better at monetizing your clips. Will this be a good thing for society? Will it make audiences smarter, more informed?

• 39:57Ed Elson?

  No, quite the opposite. And perhaps we can dig into those implications another time. But for now, let's recognize what is happening here. The future of media, whether we like it or not, is in the clips.

• 40:13Ed Elson?

  Okay, that's it for today. This episode was produced by Claire Miller and Alison Weiss, edited by Joel Patterson and engineered by Benjamin Spencer. Our video editor is Brad Williams. Our research team is Dan Shalon, Isabella Kinsel, Chris No Donahue and Mia Silverio. And our social producer is Jake McPherson. Thank you for listening to Prof G Markets from Prof G Media. If you liked what you heard, give us a follow. I'm Ed Elson. I will see you tomorrow.

• 40:41Speaker 12

  Now at Kroger, your points add up to more ways to save every trip. Use them your way to take dollars off groceries at checkout or save at the pump. You can even save up to $35 on your next fill-up. Same points, more choices. And with low prices on fresh, quality food, plus digital and personalized offers tailored to how you shop, your dollar goes further. Redeem your points online or in the app. Kroger, fresh for everyone. Restrictions apply. See site or store for details.

• 41:12Speaker 13

  Close your eyes, exhale, feel your body relax and let go of whatever you're carrying today. Well, I'm letting go of the worry that I wouldn't get my new contacts in time for this class. I got them delivered free from 1-800-CONTACTS. Oh my gosh, they're so fast. And breathe. Oh, sorry. I almost couldn't breathe when I saw the discount they gave me on my first order. Oh, sorry. Namaste. Visit 1-800-CONTACTS.com today to save on your first order.

• 41:40Speaker 01

  1-800-CONTACTS.

• 41:43Speaker 14

  Booking.com is the easiest way from a day surrounded by noise.

• 41:55Speaker 14

  To a stay. Surrounded by nature.

• 42:02Speaker 14

  That's nice. Go on, book it. It's easy. Booking.com. Booking dot yeah.