Prof G Markets - Bonds Are Warning Of A Global Inflation Crisis

Episode Date: September 3, 2026

Ed Elson is joined by John Mowrey to break down why bond yields have been rising and what it would mean for investors if they continue to increase. Then, Alex Heath returns to unpack his interview with Sam Altman. Finally, Ed shares his thoughts on why bond investors are fed up with the Trump administration.  John Mowrey is the Chief Investment Officer at NFJ Investment Group. Alex Heath is the author of the Sources newsletter and host of the Sources podcast. 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 [e-mail] Learn more about your ad choices. Visit podcastchoices.com/adchoices

Support for the show comes from alumni ventures. We've said it before. Diversification is our Kevlar vests. But do you know how concentrated your public stock fund actually is? A handful of mega-cap stocks with an AI story are doing most of the work. That's not the diversification you think, especially when so much value creation is happening before companies ever go public. Alumni Ventures ranked a top 20 U.S. venture firm by both Time and CB Insights,

gives accredited investors a smart, simple way to assemble a Bluchib Venture portfolio and participate in the value creation that is occurring in private markets. You can sign up for free, see the weekly deal flow, and decide for yourself. Or write one check into the Alumni Ventures Foundation Fund for a diversified portfolio of about 25 private venture deals. Go to AV.V.V.c. slash Prop G. Not an offer to sell securities. Venture capital involves substantial risk, including loss of capital invested. See disclosures and fund offering materials for more information. Support for the show comes from VCX.

for private tech. The U.S. stock market started history's greatest wave of wealth creation, from factory workers in Detroit to farmers in Omaha. Anyone can own a piece of the great American companies. But today, our most innovative companies are staying private longer, which means every day Americans are missing out. Until now. Introducing VCX, a public ticker for private tech, now available wherever you buy stocks. Visit getvcx.com for more info. That's getvcx.com. Carefully consider the investment material before investing, including objectives, risks, charges, and expenses, this and other information can be found in the fund's prospectus at getvcx.com. This is a paid sponsorship.

Two and five Canadians will hear the words, you have cancer. That's why every step and dollar raised matters. On September 19th, join thousands in Toronto for the Princess Margaret Cancer Foundation walk. Challenge yourself, friends, and family to walk 21 kilometers in support of life-saving research. Together, we can carry the fire and help create a world free from the fear of cancer. Register today at pmcf walk.ca.com. Money markets left. If money is evil, then that building is hell. The show goes up. Welcome to Prof. G. Markets. I'm Ed Elson. It is September 3rd. Let's check in on yesterday's Market Vitals. The major indices rose, halting a sales. Deloff. Treasury yields remained at multi-year highs. More on that in just a moment. Brent Crude's

rally slowed, but it did stay above $90 per barrel. And finally, Dell shares soared 16% after posting record revenue due to AI server demand. Okay, what else is happening? Around the world, bond markets are having their worst stretch in years. Japan's 10-year yield hit 3% for the first time in three decades. Germany's is at its highest since 2011. Francis is at its highest since 2008. British 30-year borrowing costs are back at levels last seen in the 1990s, and the US 30-year yield recently hit its highest level since before the financial crisis. This global sell-off reflects the countless worries that investors are now forced to reckon with, including climbing energy prices and hot inflation due to the war in Iran, unsustainable levels of

government debt, hawkish sentiment from the Federal Reserve, and also the enormous amounts of debt that is now being issued to fund the AI buildout. As we discussed earlier this week, Secretary Scott Bessent tried to bring yields down and failed. And so the big question for investors is the following. What will happen if yields keep rising? Hit Help answer this question. We are speaking with John Maori, Chief Investment Officer at NFJ Investment Group. John, it's great to see you again. I'll just start with kind of a broad question, which is when you look at the bond markets right now, when you look at this sell-off, which has continued into the week, what do you think the bond

market is really trying to tell us right now? Ed, great to see you. So I think there's really, you know, there's two scoreboards. There's the equity market and there's the bond market. And they both have interactions on the bond market side. There's a quick concern about deficits and debt. But what I would say is that, you know, the deficits and debt levels were pretty high 15 years ago as well relative to nominal GDP across Switzerland, Japan, the U.S. I think the real difference today is there's been a regime shift because for so long, investors thought inflation was dead.

And globalization really had kind of quenched that. And today we have a shift that's going on at. And I think we've moved from the cheapest being the most important. to the safest. And that's not that it's a binary situation, but there's a continuum. And I think that as that trickles into supply chains, you know, inflation is part of kind of the new normal a little bit, because it's almost like we skipped the insurance premium for getting everything so cheap in years past. To what extent does the war in Iran have a role to play in this new normal of higher elevated inflation. Is it that or is it multiple things at the same time? Well, I think the war in Iran

is one key component of it. I mean, energy prices are definitely affecting so many components of the economy. And there's something that touches the consumer very directly. But, you know, the high oil prices is just one part of this. You know, again, I'd kind of go back to the reordering of supply chains. Ed. I mean, if you think about, you know, we used to rely heavily on China for cheap exports, cheap labor. That was basically importing labor deflation to the U.S. So there's a lot of things that are pulling up inflation today. Oil is definitely part of that. I think the challenge for investors and for the Fed is raising rates won't necessarily fix the straight-of-hormuz. It doesn't necessarily fix the supply chains in China. So there's no doubt that what's going

and Iran is pushing inflation higher, but I think it's a bigger phenomenon that's going on when you look at bond yields across the globe. We look at the prediction markets as now a 77 percent chance of a rate hike this year. We're getting increasingly hawkish sentiment from Kevin Walsh and the Federal Reserve. Is your expectation that he's going to raise rates and is the plan to essentially try to get to 2 percent? It seems kind of crazy at this point. We've been so far from the target for so long, but are we actually going to continue to try to get to that Federal Reserve target?

You know, it's a great question. You know, the bond markets have done a lot of tightening for the Fed. I mean, you know, you had negative rates, right, back in COVID. Now you've got the highest rates going back to, in some cases, like in Japan since 96, as you cited. So, you know, what I would say about, you know, the Fed's decision coming up with the rates is, yes, they are in a tough spot because the two-year bond yield is roughly 60, ages points ahead of the Fed funds rate. So that's a signal for them. But I think they know the complication. And that is that if they raise rates, that is going to definitely increase the cost of capital. But the cost of capital is already up. I mean, everyone knows what the mortgage rates are.

Everyone knows what it costs to go, you know, buy a new car today. So you already have cost of capital higher. I think the real fascinating dynamic in the market is cost of capital is up, and that's a headwind. But earnings, resiliency. and earnings growth is a tailwind. And right now, the earnings growth is beating the headwind, if you will, from the cost of capital. So, you know, it's plausible for sure that they could raise rates. But my expectation would be that it's going to be a tough outcome for them because I don't think that's actually going to fix the problem.

And maybe if I can share one other observation, you think back to where rates were negative. And I just want to make kind of a valuation comment around this. I don't think people fully appreciated what they were paying. We just went through a bond bubble bursting. Okay. When you have the 10-year bond yield at 50 basis points, that's like paying 200 times cash flow for that asset. So people were paying enormous multiples for bonds.

And so what we've really done is normalize the bond market. Everyone thought it was going to stay that way forever. But inflation has kind of woken up. And it's really re-rated bond yields back to levels that are really more normal if you look at history. I mean, you say that basically Kevin Walsh is trying to fix this, but he might not be able to fix it. There's not much that you can do at the Federal Reserve to fix what's happening in Iran. But Scott Besson has tried to fix it, and he tried to fix it with buybacks, and it didn't really work. or maybe it did for a couple of days and then it didn't.

What do you make of the buyback strategy as a means to lower borrowing costs in America? Is it the right strategy? Is it the wrong strategy? What do you think of it? Treasury's job is to finance America. But there's no rulebook on where on the curve the Treasury can issue that debt. It's a really fascinating topic. How many 30-year bonds is the right number?

How many 10-year? What the Treasury is effectively doing is they're saying, okay, we would rather finance more of America's debt at the short end of the curve. So it's basically like taking out an arm on America's debt, an adjustable rate mortgage, so you have to refinance more versus the 30-year fixed. So it's their prerogative.

They can do it. I think that it's a relatively small compared to what was Operation Twist back in 2011 when the Fed stepped in and said, hey, we're going to buy bonds back on the long end and sell short term. So it's a relatively small move from $2 to $4 billion when you think about the total size of the balance sheet of the U.S. government. But it's in his prerogative to do so. He's getting criticized for it.

But I guess I would just kind of throw out again kind of the thought piece on what's the right number for 30-year bonds that the U.S. government should. hold. And whenever he decides he's going to go out and buy back those bonds, what he's effectively doing is he's buying those back and he is exchanging what was at a lower interest rate for something that's at a higher interest rate. So that could be good if rates fall, but it looks a little bit like active management to me, which is an interesting role for the Treasury to be in. You mentioned earlier that the debt situation in America and kind of across the rest of the advanced world isn't that much different.

compared to what it was on a nominal basis, or compared to GDP, at least. I wonder if perhaps part of the concern for investors right now is that it seems like the longer this drags on, the clearer it becomes that our government just doesn't really care about fiscal responsibility. Like if we get to the higher numbers we get to, now we're at $40 trillion in U.S. national debt, the longer this goes on, the more we start to realize, okay, no one's going to fix this. This is never going to change. And perhaps that might be the concern that is being priced in. Is that a concern to you? Do you see our debt and our rising deficit under this administration? After he said that this was an important thing to balance the budget,

at least, that was part of his platform, is that a real concern for you? No doubt. I mean, look, I mean, the debt continues to expand. But if you think about how we deal with debt in kind of the, you know, the modern economy, it's through inflation. I mean, the reality is that we repay our debt with many dollars because the dollar and all currencies continues to depreciate over time. And everyone understands this. Everyone knows what a cup of coffee cost 10 years ago, 20 years ago, 30 years ago. So the way that we deal with this is inflation. and it's the most insidious tax on the global population that's ever been invented.

And you can see it very slowly. And it's a regressive tax ed. So it taxes everyone. So are we going to be able to contain it? I mean, if you look at the kegher of the national debt, it's about 7.6%. And that kegher is pretty consistent. So we like to spend money. And unfortunately, when you don't pull enough in taxes and you spend more than you bring in,

you've got to issue debt and the markets tolerate it. And particularly in the U.S. they tolerate it because we're the strongest, most liquid, deepest market in the world. But the way that it's paid for is through inflation. And I think that for years at it was masked by globalization. And now that that's being reordered, I think that it's revealing that, hey, maybe that was temporary. Maybe inflation is permanently in the system. And it's a byproduct of what we've done fiscally and monetarily, but it's also the price that you should pay for convenience. I mean, if you look at, again, going back to what you said about what's going on in Iran,

yeah, the oil markets are probably going to price in a new risk premium. Because it's like, hey, if a strike gets shut down, that's a big problem. And when you have a just-in-time supply chain where everyone was used to getting things just when they needed them for their, you know, distributors and for their production, everyone got used to no inflation. I think that's what created kind of the bond bubble. You have no inflation. You can spend more than you need, and you can issue debt very cheaply.

And now we are having to pay that bill. And what's fascinating, though, is if you told me that rates on JGBs would go from negative to three over the course of three or four years, I would have thought the markets would have had a really tough time. The NK is at all-time highs. The S&P's at all-time highs. Why is that? It's because earnings are overpowering what's going on with the cost of capital. And so I think the Fed needs to pay close attention to that because at the extent they raise rates

and at the extent they slow down growth in the economy, it could be a much more challenging situation because it's ultimately the U.S. companies that employ all the people. Say yields continue to rise even higher. They've been rising all week, practically a month. Say it continues. What does that mean for investors? then what does it mean for everyday Americans? Well, I think that if you look at the continual rise in the cost of capital, there's no question that that is, again, a regressive tax, if you will,

because that's affecting everyone up and down economic scale. In terms of how it affects the market, though, Ed, it's really going to come down to when does the cost of capital start to pull away the earnings resilience? And the reason that the market has tolerated higher rates is because we're probably sitting inside the biggest CAPEX cycle in history. And so that's trickling to so many areas of the economy. And it's absolutely tied to the AI buildout. And so that is allowed the economy to tolerate higher rates. So no one knows exactly what that number will be when that starts to kind of erode the durability of earnings.

But as long as we continue in this CAPEX cycle, then that's going to allow earnings to be resilient. and that's going to allow the markets to be resilient. And so that's really the tug of war right now between cost of capital and earnings resiliency. And to be honest, you know, the way you deal with it is you invest because you want to be on the side of the companies that can pass through that inflation. That's really the best gift that Americans have. We have a very open and liberalized market. And anybody can do it.

My kids can do it. You can open an account. You can stick in $500 when you get that saved up. You can start it. And I've done that with my kids. I've started investment account. I've shown them when it's up and down. So I think that that is how Americans should try to tackle, again, going back to the most insidious tax ever put on on the global economy, which is inflation.

You need to be an investor and get on the right side of the equation. John Murray is Chief Investment Officer at NFJ Investment Group. John's always good to see you. Thank you for your time. Thank you, Ed. After the break, an inside look at OpenAI. Support for the show comes from Vanta. When you run security for a company that's scaling fast, the stakes just keep climbing.

More compliance frameworks, more vendors, more risk, and a board that wants to see it all in one place. But your compliance data can be all over the place, controls in one tool, vendor risk in another, customer commitments buried in contracts. Your TMS has to spend more time stitching it all together and less time running the program. And every quarter, you're basing decisions on last quarter's data. Vanta connects it all. Its market-leading agentic trust platform is built for enterprise scale with 400 plus integration. and continuous monitoring. Vanta automates evidence collection, surfaces the risk that matter, and organizes your audit around your auditor's requirements, and that Vanta agent works around the

clock from full program contacts, so nothing slips through the cracks. The numbers back this up. Vanta delivers a 526% return on investment over three years and pays for itself in three months. Vantage is trusted by more than 16,000 companies, including Snowflake, Atlassian, and Ramp. Learn more at vanta.com slash markets. That's spelled VANTA.com.com markets. Two and five Canadians will hear the words you have cancer. That's why every step and dollar raised matters. On September 19th, join thousands in Toronto for the Princess Margaret Cancer Foundation Walk. Challenge yourself, friends, and family to walk 21 kilometers in support of life-saving research. Together, we can carry the fire and help create a world free from the fear of cancer. Register today at

pmcf walk.ca.ca. So like any good millennial, I have a love-hate relationship with Gen Z. It's the phenomenon rattling millennials. They just look at you. They want something bigger themselves, lifestyle is a priority, motivation is being inspired. But regardless of how you feel about Gen Z, it's undeniable that they're changing national politics. Generation Z is increasingly showing less loyalty to traditional political parties, many now more

likely to identify as independent. So what is going on with the kids? I think the biggest misconception about Gen Z's politics right now is that all of a sudden they're all socialist. That is just not the case. They are embracing candidates who are offering new bold ideas in the absence of those ideas from establishment Democrats. This week on America actually, Gen Z researcher Rachel Janfaza joins us to separate

Gen Z fact versus fiction. It's not rocket science. And this is, you know, I keep saying like, young voters aren't that complicated after After all, it's pretty simple. Catch us every Saturday on YouTube or wherever you get your podcast. We're back with Profi Markets. OpenAI has had a tumultuous few weeks. As we discussed a couple of days ago, roughly 700 rogue open AI agents hacked the machine

learning platform hugging face in July, and reportedly OpenAI didn't know about it until later. Meanwhile, the company is facing multiple legal battles. On Monday, Apple escalated. its trade secrets lawsuit against the company, telling a federal judge that OpenAI is actively destroying evidence. Meanwhile, 30 new lawsuits were filed this week over a school shooting in British Columbia, which killed eight people. The suits alleged that Open AI could have helped to prevent this, as the suspected shooter had heavily interacted with ChatGBT. This all comes amid the anticipated

release of Open AI's newest model Astra, which is reportedly the first to have critical. cyber abilities. So to get into all of this, OpenAI News, we are speaking with Alex Heath, who just sat down for multiple one-on-one interviews with the CEO of OpenAI, Sam Altman. He also spoke to more than 20 people at the company for a cover story. He just wrote for Time Magazine. Few people know more about what's going on at OpenAI than Alex. Alex is the author of the Sources newsletter, and he just launched the new Sources podcast. Alex, thank you for joining us. good to see you. Let's just start with your interview with Sam, which was just released this week. I watched the interview, fascinating on many levels. We'll get into it. But what were your top

takeaways? Oh, man. Thank you, Ed. I appreciate that. Yeah, I've been spending a lot of time at Open AI and with Sam, and we covered a lot of ground in the conversation. A couple things stood out. They are really slowing down frontier research in a way that is unprecedented in the industry speaks to the capabilities they're seeing that have not yet been released. They have Astra, their next family of models coming, which I've gotten to see as part of my reporting process for the time cover story you were talking about, but no one externally has used yet. And what they saw, and I was actually in the office as this was happening, was even further frontier training runs that they have after Astra were showing, as Sam put it to me,

various degrees of misalignment, which is AI speak for the AI is not doing what it's supposed to. And you see that with hugging face, right? The hack on hugging face that everyone knows about. But I think it speaks to as these models become more agentic, doing things in browsers, taking over computers,

the security risks go way up. And OpenAI's research team decided, which that company is still very much a research culture-led company, Even with all the commercial stuff that we'll get into, that team has pushed them to slow down, and they are. That said, they're also in this fierce model race. They want to get asked out the door as soon as possible. I think it's going to be any day now. It's going to be a very powerful, capable model.

And then we touched on a bunch of stuff. We touched on recent leadership departures. We touched on the IPO, Anthropic competition, the device work with Johnny Ive. He told me they're going to make a humanoid robot. There's a lot to go over. You asked him about the hugging face incident and the extent to which it was a security breach, an accident. Here's what he had to say in response.

It's a safety failure for sure. There's a question of how much you're supposed to understand that there's a security issue or alignment issue. I think it's mostly been reported on as a security issue. I think I understand it personally more as an alignment issue. But in any case, yes, that was a bad thing. And I don't want us to make excuses for that because I don't believe that's how we fix it. the more we're like, oh, our nice little model, he would never do anything bad.

Like, you know, it was just a little e-vals, harness, misconfiguration, no problem, nice little model. That would be a very, if I said something like that, then I think you should be like, oh, this is really bad. It seemed as though his view is we're recognizing the problem, we've admitted the problem, so it's not that much of a problem. What did you make of his comments? I can see that take. I think he was throwing a little shade at other labs who, you know, hype up the capabilities of the models and then when stuff happens, don't really talk about it or kind of throw it under the rug.

I think that was more of the meaning behind that statement he was talking when he was explaining that. And, you know, there's another part of the interview he's like, people call me the yellow CEO. He was referring to like something that Dario Amade said. at a conference last year. Yeah. And... Which you acknowledged, and then he sort of refused to admit that that's what he was saying,

which I also thought was interesting. I won't acknowledge or deny or not deny who I'm talking about. And you're like, yeah, you're talking about him. Well, I look, Sam and Dario, I think, live rent-free in each other's heads always. It's very apparent. But I think, you know, I'm not saying this for a fact, but I think, If you were Sam, Anthropics about to have maybe the biggest IPO of all time in the coming weeks, if I'm Sam and you're slowing down Frontier Research for a good reason because you have concerns about the alignment risks,

it's probably like a good PR move to go out there and be the safety company, right? Because that's what Anthropics' whole claim to fame is. I mean, that's how the company began was a bunch of safety people left Open AI, right? So I think there's a genuineness in that open ad really does see capabilities internally that frighten them in terms of their ability to align them. And I also think, look, it's convenient timing. I think both can be true. Something I've been wondering about this hugging face incident, which I feel like is people weren't talking about and then suddenly everyone's talking about it. To me, I can't tell whether maybe they're proud of it because it's this example of their,

agents being very capable and going out there and doing something that is very sci-fi-e. And then also they get to say like, oh, but we're addressing the problem. And part of me wonders, I mean, on the one hand, I feel like maybe they want us to see it as evidence of how capable the agents are. But then part of me thinks, the cynical part of me, thinks, well, maybe you guys just aren't really doing a good job with your own security. And it doesn't say that much about the advanced capabilities of the models, as much as it says about that you're a kind of scrappy startup and you're not doing your job when it comes

to cybersecurity. How did you see that incident? Do you see it as a serious statement? Does it say something big about where AI is headed right now? I think it does. I would encourage everyone listening to watch Open AI's Black Cat presentation where there are researchers, I cite it in the interview with Sam, where the researchers walk through how the model actually escaped and hacked hugging face. It's remarkable. And it's the first incident of this happening. And at the same time, I think you're right to point out, opening I didn't have the safeguards in place. If this is a company, one of the two main companies were trusting with the future of AI, upending everything, how should we feel about the fact they didn't have the safeguards in place

to keep this from happening? They didn't foresee what would happen. And I talked to, you know, the chief scientist and head of research and a bunch of people. head of alignment about this. They had safeguards that they had developed that would have stopped this. They didn't implement them for the training run that led to the hugging face breach

because they didn't think the model was going to be that capable. So there is a sense of what's happening inside these labs at the frontier is the model capability is getting ahead of the human researcher's ability to foresee what's happening, which is you get into this concept

of an AI recursive self-improvement, RSI, which Sam and I also. also talk about in the interview, particularly around the IPO, which was, I thought, interesting, the way he framed it around the IPO. People in the frontier labs really feel like they're on the edge of like a precipice that's like potentially amazing and also cataclysmic and the sense of AI that can build itself recursively self-improve, not need a human in the loop, even the chief scientist of open AI. And that's already starting to happen. People talk about RSI as a thing that hasn't happened yet, it's definitely starting to happen. And so then you get into this AI, people

talk about this concept of takeoff and, like, there's no going back. It can get very spooky, very sci-fi, you know? I mean, Sam said that in the interview too. It feels like sci-fi. And it kind of is, like when you see what happened with Hugging Face. At the same time, to your point, yes, open AI should have had safeguards in place. One of the big themes we talk about on this show is the profitability of AI or the lack thereof among the AI Open Air, I included. You guys talked a little bit about the AI buildout and the extent to which it is or isn't profitable.

I thought his comments were really interesting. Here's what he said. Not worried about our compute buildout plans. I am worried about the world's compute buildout plans. I think we are going to be able to use all of the compute very profitably that we were planning to build. But I am seeing the first signs of what feels to me like unsustainable silliness of, you know,

random new neocloud popping up, people claiming that they're going to build gigantic amounts of compute next year that I think they don't have the revenue to support or a buyer. Yeah, I definitely feel like some fear about what the world is doing as a whole, although I think we feel very good about what we've committed to. What were your reactions to those comments? Oh, it's interesting, right? I mean, there are a bunch of neoclots that are making gigantic contractual commitments. And that was the position open AI was in last year, right? With Stargate and Sam with Larry Ellison,

and in the White House, you know, these trillion-dollar headlines that were flying around. And you could look at a lot of companies making huge commitments and just kind of do basic napkin math of what revenue do they have

to support those commitments. Open out, obviously, it's not public yet. We don't have an S-1. I, Ed, I got to be honest, I know you've had people on the show that are very critical of the finances and you have been too, and I get that. Until we see it, right,

you kind of have to be. It's your job. I've walked away from the reporting process and I talked to Sarah Fryer, I talked to all the compute leads, I talked to everyone, feeling a lot more confident that this isn't as fragile

as maybe people think it is, at least with Open AI and Anthropic. I mean, Open AI is at about a 40 billion run rate, enterprises past consumer in terms of the revenue mix. The most interesting part about open eyes business from the interview that I actually wanted to talk about with you

was A, that his comment about the Neo-Clouds, but B, there was a moment where I was asking him about ChatGPT growth because they hit a billion users recently. But it took a long time. It took, there was an incredible ramp up to hundreds of millions of users and then they teetered on a billion for months

and months and months. And I thought that was unusual and suggested maybe more competition in the market, you know, his chat losing its luster. And he very matter of fact, it was like, well, we decided to put all our compute into codex, in the coding to compete with Anthropic. And I was like, wait,

so your revenue is a direct function of where you can put the compute? And he's like, yeah, completely. So they're kind of like these labs are like alchemists in a sense of how they control their business because they are so compute constrained that it's not like a normal business where, you know, there's demand that you can forecast and control. They're just, they don't have enough. They don't have enough compute to serve the demand at any given point. And so Open I made a decision to shift compute that then slowed its consumer growth to grow its enterprise growth.

So yeah, these labs are a little bit like, Alchemists right now. And I don't think that's, I don't know if there's precedent in the history of modern capitalism, especially for two companies that are about to have trillion plus IPOs. Like, I don't think there's precedent for that. And I found that very interesting. What parts of your interactions with particularly Sarah Fryer made you feel more confident in the financial sustainability and the financial health of this company? Well, look, they're not profitable. I think if they stopped training today, they could be. I think that's been the case, by the way, for a while. What happens to the multiple on the company,

on the valuation if they stop training, if there's no more frontier to chase, that's a bigger question, obviously. The economics of just like inference are actually really good. The frontier labs operate at incredible margins on inference. Inference meaning running the models themselves that people are using versus trying to build

these incredible frontier models for the future. Correct. And if you look at what OpenEI is done with token efficiency driving the cost down of tokens, even for the sole family, most recent one with Luna and the smaller ones. This is not the behavior of a company that is teetering. And they're building a real machine in-house that is, you know, they're still growing up.

They're still a startup even though they're huge. Like the culture is a startup, right? Like that company operates on like a 12-hour horizon at any given point. I can just tell you from being on there for a couple weeks. At the same time, like, they're starting to become more regimented about spending. and the inference business is going to be very good. I think. I've no doubt about it. Now look, like, if training continues to balloon,

maybe that's an issue. I think a lot of people in AI think RSI will lead to training costs declining. Also, like, the chip gains that they're making with their custom chip, jalapeno, like lowering the cost of compute is a big focus for them. And I think they see a line of sight to it. So, look, it's unprofitable. I think that's a choice. I don't think that's because they have no choice.

But, you know, we'll see. We don't have an S-1 yet. Just to wrap up here, you asked him about the growing anti-AI sentiment and the data center backlash. I'm not sure he gave you much on how he actually felt about that. What do you think he thinks about the fact that so many people hate AI and not just AI, but open AI. I was really curious to hear

how he thought about this, especially with, you know, the recent attacks on his home personally, right? And the data center

backlash is stronger than ever. I thought his, his response to me was, you know, it was interesting. It was basically, like,

if we make a better product, they'll like it. And, you know, I kind of respect it in the sense that it puts the pressure on them to deliver.

You know, I've had some pretty magical moments. I'm not going to lie, with AI. in the last few weeks chatypity work club co-work

doing basic stuff I mentioned in the interview but like filling out a post office web form so a package gets picked up and it would have taken me 20 minutes

I just did a prompt and it did it and I put the package out and it was gone the next day. Little stuff but it starts to add up and you're like

wow this is like this is a nice value like I'll pay $100 a month for this. Most people do not use chat that way they use the basic free model it's better

Google search it can do some other things but it's like people look at it as like insanely expensive Google search that's like detrimental to the environment. That's how most people think about AI. And I think what he was trying to say there is like as we move into this agent world, which like Astra the next model, it's trained to use enterprise software faster than humans. So like as we move into this world, we're like, you hate using your

expense software and you just tell chat to go do it for you. Like maybe you start to feel differently about AI if it's like actually making a meaningful positive change in your life. And I think that's what he's banking will happen. That's a, that's a prediction by him. I'm not saying that's going to happen. But that was his answer. And yeah, I don't know. I don't know if people will find that answer fulfilling or not. I think people are going to keep hating AI. Alex Heath is the author of The Sources Newsletter and a host of the Sources podcast. Alex, appreciate your time. Thank you. Always. Thanks, Ed. Circling back to our discussion with John, it appears.

that the bond markets have finally reached the end of their line with the administration. Turns out that if you drive up inflation with tariffs that ended up just being returned to corporations, and if you double that inflation with another forever war that just so happened to choke off a fifth of global oil supplies, and if you rack up two trillion dollars in annual deficits, and if you explode our total debt to $40 trillion, dollars. And if you show the world that you don't really care about any of this, nor do you think that any of it even matters, then eventually, yeah, investors will price that in. The yield on the 10 year has risen to more than 4.8 percent, its highest level since 2023.

The yield on the 30 year recently hit 5.3 percent, its highest level since the financial crisis, and the yields on foreign bonds from Japan to the UK to Germany and France have all hit multi-decade highs. Borrowing costs around the world are soaring as the bond market speaks out about an impending inflation crisis. Now, keep in mind, this is all in spite of the fact that Scott Bessent used the Treasury Department to try to muzzle the bond market. He issued billions of dollars in long-dated bond buybacks. that was his attempt to fix the price of U.S. bonds, to bring yields down, and, as his former colleague, Stanley Druckenmiller put it, to prevent the bond market from speaking. But you can only shut investors up for so long. And in his case, it was a couple of days. Eventually, reality has to catch up. Now, has the full force of our fiscal reality and our inflationary reality fully set into the markets yet,

I think probably not, but it appears that it is beginning to. Whatever appetite investors used to have for this administration's BS is clearly beginning to wane. They are fed up with it. And if you can't tell from the sound of my voice, I am fed up with it too. Okay, that's it for today. This episode was produced by Claire Miller and Alison Weiss and engineered by Benjamin Spencer. Our video editor is Brad Williams. our research team is Dan Shalon, Kristen O'Donohue, and Mia Silverio, and our social producer is Jake McPherson.

Thank you for listening to Property Markets from Profiting Media. If you liked what you heard, give us a follow. I'm Ed Elson. Tune in tomorrow for our conversation with Tyler Cowan. Two and five Canadians will hear the words, you have cancer. That's why every step and dollar raised matters. On September 19th, joined thousands in Toronto for the Princess Margaret Cancer Foundation Wall. Challenge yourself, friends, and family to walk 21 kilometers in support of life-saving research.

Together, we can carry the fire and help create a world free from the fear of cancer. Register today at pmcfwalk.ca.ca.ca.