Fed Split, Oil Surge & AI War: Why Markets Are On Edge Right Now

Hello and welcome back to the Market Maker podcast. It is Thursday morning on the 30th of April. We are recording this, so still a day to go, or in fact 2 days if this is the morning still of the market week. However, so much has happened already. We're still awaiting in a few hours' time the likes of the European Central Bank, the Bank of England to come out and announce their latest policy. But in this conversation, we're going to talk about the Fed. Of course, it was Jerome Powell's last one before he departs, but he's not really fully party. Uh, we'll get into that in a moment, and then we'll talk a little bit about oil going back up to $120. Latest rhetoric out of Trump making investors nervous and seeing the global benchmark moving up north of 7% in trading on Wednesday. We had a deluge of big aftermarket tech earnings. Google, big standout winner, up 6%. Mirror image opposite of that was Meta, down 6%. So We'll have a look at how they are faring under these enormous amounts of CapEx at the moment. And then the AI race, there's some winners and losers, and as you can imagine, two big names, OpenAI and Anthropic dominating the news flow this week. So perhaps we could start with the Fed, where the US central bank kept rates unchanged as 100% priced by the market. I did look at that. Before we were going in, because we were talking to a group of students last week. And so they kept rates at 3.5% to 3.75%. It was the third consecutive meeting with no rate move, but very fractured board, it would seem, a committee. Tell me a little bit more, Piers, about the split, and have you ever seen anything like that in your time in markets?

Well, this is it. So When nothing happens at a top level, like when a central bank does not change interest rates, then you've really got to kind of delve into the finer detail to really get any sense for the direction of travel, right? So it becomes, well, what's going to happen to interest rates at the next meeting? Or obviously, well, the next meeting will be, actually, that's going to be, let me just think, might do some maths. That's the start of June, isn't it? So we've got one more meeting in the first half of this year, right? And then we've got second half of the year, we'll have another 3 meetings. And so it's kind of thinking about, right, what can we get out of this event when nothing's happened on the headline? What can we get out of it to give us more insight about where rates might go in the future? So we look at the vote split. So this is where you get an FOMC committee, the Federal Open Market Committee. That's the selection of Fed governors that are on this rotating system, but basically you get these 12 people that are essentially voting on what to do with rates. Now, normally, I mean, historically the Fed's been famously, I would say, unanimous on its decision-making. I mean, they, I would hope, have a healthy conversation and meeting where different thoughts and angles and ideas are brought, but in the end they tend to come together. And in unison deliver their announcement. They're going to hike, they're going to cut, they're going to do nothing, right? But since Trump's come along and really changed the game, I would say, and he's installed this, this Myron onto the FOMC, which is really Trump's puppet, it's kind of set the cat amongst the pigeons. And I would say ever since then we've seen this split that's kind of come into the situation. So you've got the Trump factor. You've then also got the big problem with what's going to happen with oil prices and how's it going to have an impact on inflation. And whilst on the one hand you've got the idea, there's three prongs to this in my mind. There's, if it's just a supply shock, an energy supply shock, well, interest rates can't play a role there, right? So you shouldn't be changing interest rates if it's purely an energy supply problem. The problem comes if that energy supply shock filters into other parts of the system. And we've talked about this at length on this podcast. We're not going to go into the details again, but you know, you start to get the price. One good example is the price of fertilizer goes through the roof. Okay. And therefore, right, that can spill into food inflation and so on. Right. So it will inflation filter into the rest of the economy and will it lead to longer-term inflation expectations going up. If that's the case, then the, the central bank should act and they should be hiking. Okay. The third kind of leg of the stool here in terms of your decision-making process is then, well, okay, well, if energy prices are so high, well, isn't that going to have a negative impact on the economy then? And isn't the economy going to slow? In which case, actually, Maybe we shouldn't be hiking, maybe we should be cutting. So you kind of got this real conundrum in the mix. So whilst Trump is an easy, I would say lazy reason to pin this sort of this fracturing of the committee and the vote split, you know, I think ultimately it is a really difficult moment in time for this committee to decide what to do. So you've got Myron who wants to cut. He's been asking for cuts ever since Trump shoehorned him into the system. So that hasn't changed. That's not a surprise. So I guess we should talk about the other 3 people and that they did not dissent against the idea of keeping rates on hold. So this gets even more kind of nuanced now. Okay. They absolutely voted to keep rates unchanged. There was only one person on the committee that was different, and that was Myron, who wanted a cut. The other 11 unchanged, right? So then it becomes about the statement that they release, which is just a kind of short, I mean, what is it like two sides of A4 where they are ultimately, the Fed are ultimately conveying their current sort of opinions on the economic situation and therefore justifying, you know, why they have made the decisions that they have and then really kind of just giving a sense for what might be coming next. And it was that what might be coming next part that caused some disagreement around the table. So the three that dissented, so here's some names for you: Beth Hammock, who's the Cleveland Fed governor, Lorie Logan, Dallas, and Neel Kashkari, Minneapolis. Those three voted against the statement. Now, so basically in the statement, they disagreed with keeping the easing bias language. So basically the statement that's been rolled out for, for many meetings now has had language in it that has been suggesting that the next move in interest rates would be downwards. So these three, this is a hawkish development. These three have now dissented against that. They want that language changed, implying that they think actually the next move on rates may well be up. Not down.

Is this not just a sensible, logical way of policy evolving with what's happening as you've explained? Because here it's not, as you were saying, it's not like a, I'm jumping straight into a hike camp. It's a subtlety and it's a minority moving across while the consensus is still hold. It almost feels like a moderate hawkish hold. Rather than an aggressive hawkish hold.

Yeah, this is it. I think the media like to jump on this a lot. I mean, you are right. If this didn't happen, if this wasn't happening, I'd be worried. I think this is actually great evidence of a really healthy functioning decision-making body. That's how I see it. So I take a lot of confidence actually from reading this kind of stuff. So yeah, but Ultimately, there has been a shift. It should surprise absolutely no one on the planet. And yeah, I would say we're moving a little more toward, well, you could say it's hawkish. I'd say there's another kind of part to this though, which is about Powell himself. Whilst his words at the end of the press conference, because this is now that Trump's nominee Warsh has finally been voted in by the Senate, so he's been confirmed now as the next Fed Chair. So we will see Warsh at the table at the press conference next time round. Powell's words at the end of his final ever press conference as Fed Chair, and it's been 8 years that he's been sat at that seat, he said, I will not see you next time. And then he strolled out, but he's not going far. This is the thing. Because, well, for those who don't remember, there's been a bit of a very late final kind of warsh voting through here. We've been, the Senate have been blocking it. This is all to do with Trump and Trump kind of instigating a criminal investigation against Powell and how the refurbishment of the Fed building, which went a billion plus over budget. Trump's trying to get in there and claim that there's some kind of criminal shenanigans going on. And so there was an investigation set, all right? And one of the senators, I forget who his name was, a Republican senator, was basically saying, I don't like this, Trump. This is going too far. Unless you drop that criminal investigation, I will not vote through your nominee, Walsh. And so it kind of just, everything was stuck. Now the Department of Justice probe has been dropped, which has now enabled Borscht to be voted through the Senate. But there's a problem here because Powell has said that he's made a decision to stay on as a Fed governor, right? So he, he's not gonna be Fed chair anymore, but he said, I'm staying on as a Fed governor. And he wants to kind of, he's worried about the Fed's independence basically, and wants to make sure the Fed and their decision-making process doesn't get coerced by political pressure, essentially. Now the thing about this is it's quite interesting because if Walsh now comes in as chair and Powell stays in his seat as governor, it actually pushes Myron off the FOMC committee, Myron being Trump's puppet. So actually this move by, it's quite a sort of power move by Powell actually. This will actually make the committee more hawkish going forwards because it pushes Myron out.

Yeah, I like that. That's, it's kind of like a subtle passive-aggressive move from Powell, which is the opposite style of Trump, which you know is going to wind him up to a huge degree. So yeah, no, very super interesting actually in that last point and how things are going to play out with Powell still in the mix. But well, look, let's keep the show moving and let's then talk about one of the contributing factors that you were talking about there. And that was the fact that oil prices have gone up 24% in the last 10 days is the kind of the stat line. And in the middle of this week, so on Wednesday, the president warned that the US blockade could last for months and told Tehran to capitulate on nuclear demands. So yeah, a much more aggressive stance here. There was also reports that Trump is receiving new briefings on potentially new military actions that could take place. Within the strait. So yeah, thoughts on that and those latest developments with this now highest price of oil we've had since the conflict began.

Well, that's the headline, isn't it? You know, I think, well, let me just get the actual numbers here with regards to Brent crude and it kind of, yeah, it hit 120. Where did we get up to? 125, I think it was. Or just short of that. Sorry, it was just over the 120 handle. So yes, the highest levels we've seen since this crisis began. And look, we're kind of 8 weeks in now, right? I don't know. You go back to the start of this thing 8 weeks ago and everyone's like, oh, it'll only last a couple of weeks. You know, it's not going to be a problem. Yeah, oil prices have spiked, doesn't matter, they'll come back down. And you know, we're definitely in a different scenario now. And I think ultimately the price reaction on oil overnight kind of obviously, I would say finally is properly pricing in this. And what is it pricing in? It's pricing in this being a much more sustained, you know, longer-term situation. And, you know, Trump has, yeah, basically there's lots of rhetoric, like maximum pressure rhetoric is how the media portraying this, obviously. Trump's got his blockade up and Trump's been hinting that the blockade's gonna last for months. He's told Tehran to capitulate with regards to some of their demands, you know, around the ceasefire negotiations. So yeah, I guess Trump's taking a bit more of an aggressive stance. There's some rumors that they're looking at some, you know, more sort of targeted military action against Iran, right? So it's escalating is the point. We're 8 weeks in and it's escalating. So I think coming back to those inflation concerns and will this event, A, lead to medium to long-term inflation expectations rising, causing central banks a real headache where they might have to hike, but B, how much of a negative economic impact is this all going to have? And then this is where we get this stagflation risk where you get inflation rising as economies lose momentum and worst case you get recession. Plus inflation at the same time. And that's the worst problem of all. So yeah, I mean, look, it's an ongoing, obviously it's ongoing where this thing's escalating. We're 8 weeks in and I can't see, yeah, I just can't see an end to it at this point.

Just thinking about those other episodes when Trump has kind of re-escalated and it felt like he was always going to adopt or have that card to play. Not unless the rates market started to move aggressively.

Yeah.

So at the moment, what the rates market is moving because of all of the macro fundamentals you're talking about just there more broadly with the Fed, but not the market does feel very open to overlook the historical basis. So I was looking at the S&P 500 going back to multiple different crises that have happened from a war conflict perspective from the 1970s to the current day. And within 12 months' time, the stock market on average over those, I think it was 9 conflicts. What percent do you think is up 12 months after the start of a conflict? Uh, and I, you realize I'm saying up, not down.

Yeah, yeah, yeah, yeah. It should be going down. I don't know, like it's probably over 50%.

No, it's 7. Okay.

Well, 7%, it finishes higher.

Yeah. So after a 12-month period, and in fact, effectively, if you look at it on a 1-week, 1-month, 1-quarter, 6 months, a year, it on average flips quite positive. So the point, yeah, I guess the point I'm making is it feels like the broader market, although there's a sense of fragility and lots of sell-side banks talking about this saying, you know, it's ripe for a bit of a pullback. There's a lot of the AI anxiety at the moment about CapEx and so forth, which was already simmering prior to the conflict rearing its head.

Yeah.

But for the moment, I would say broader market, it feels like he's got a little bit to run here. And I guess there's not solid concrete yet data points that have fed through saying the consumer's really bulking in terms of their consumption perspective.

Yeah. I mean, is there a whole nother layer to this that no one's talking about, which is, is this actually an economic war? With the likes of China, for example, because basically the US economy, if you take all the big economies, they're in the best position here in terms of being sheltered against the fallout effects of a higher oil price. That's because A, they don't get any oil from the Middle East at all. That doesn't mean to say their oil prices haven't gone up as well. They have. So your WTI crude is higher, but it's much less high than Brent crude. Plus they've got a bag load of emergency reserves. All right, that's only finite, but still. So, and then on top of that, you've got this AI boom going on. And actually, when you— we'll talk about earnings in a minute. I mean, the S&P's all-time highs because US companies are just absolutely smashing it. So Trump's, I think, sat there quite comfortable going, well, actually this oil price being this high, it's not actually hurting us. It's actually really hurting our competitors though. So you talk about borrowing costs and you talk about bond yields. Well, think about the US 10-year yield is at 4.4%. And actually, you know, that's, it has gone up, but you could argue it was artificially low before this conflict. I would say if the 10-year goes above, it hasn't got much more room. If it goes above 4 point, if it gets to 4.5% and higher, then I think we're into a slightly different situation because that is then you're getting feedthrough effects to US yields. But at the moment, US yields are quite stable. You go and look at some of these other ones. I mean, German yields are now the highest they've been like for decades. I'm not right. It's just kind of finally broken through. Their Bund 10-year yield is up through 3%. It hasn't been that high for decades, literally. Japanese bond yields, they're at decade highs. You go and look, it's a bit basically APAC bond yields and European bond yields, and certainly here in the UK as well. Again, record highs basically. So you've got Europe and APAC feeling the squeeze. And is this a kind of, I don't know, is this a kind of a play by Trump where as this is going on, he's realizing actually this is relatively good for us because it's worse for everyone else. And that maybe puts him in a little bit of a position of what he thinks is power and strength.

On the oil front, one of the other major headlines of the week was that the UAE announced that it's withdrawing from both OPEC and the wider OPEC+ alliance. Now, this has some significance because I think UAE's third in the pecking order as far as production values are concerned. So they're a significant contributor to OPEC. So is that going to change at all, this oil price in respect to the supply side? Because I know Abu Dhabi's been building out the infrastructure for some time. They've had strong desires to want to utilize and pump that to a greater degree. I think the margin is more like 20%. More than it had been prior conflict, but they've been shackled because Saudi Arabia's on the other side going, we want to control quite tightly the supply and the market because we need the oil price at a certain level because we've got our own funding objectives and ambitions that we need to fulfill. So the desire to pump more is always the you, you're just being put down by the bigger member of the family, so to speak. So is this, is that, is that explain the rationale or is there more to it in here? And what are, what are the consequences for, for some of the other players involved?

It explains the rationale well, I think. I mean, this is a 60-year alliance that's breaking up here. So I mean, it is, it is a landmark moment. But there's been grumblings, as you've said, for a few years where UAE haven't been happy. So you could say that it's, it's not surprising. I think the timing of it is political. And I was actually looking at a chart, I don't have it in front of me now, but looking at Iran and Iran's targeting of other Middle Eastern countries in terms of the attacks that Iran have been carrying out on infrastructure and, and so on. And actually the UAE seemed to be singled out by Iran. You are talking like 10 times the number of attacks on the UAE compared to, let's say, Saudi Arabia, or some of the other Gulf nations. So I think the UAE feel like they're being targeted. I think that's kind of pushed them into Trump's side of the camp here. Clearly this is whether they, well, I was gonna say whether they intended it or not. I mean, they have to have intended, intended this. This is basically them saying, right, we're on Team Trump.

And with that, one of the other stories from a few days ago was the UAE had a swap line request to the US.

All right.

Aimed at easing liquidity pressures on the country's banking system. That came around 3 or 4 days before they exited OPEC.

There you go.

So they're going to pump more. Trump goes maximum pressure, oil prices squeeze up, they start to ratchet the extra 20 to fill the void.

Yeah, this is important. Well, like, why? There's two questions now, now that it's happened and the news is there. It's like, can they actually produce more? Well, they are like, on a simple level, the answer's yes. And actually the interesting dynamic about OPEC is there weren't many OPEC members that had spare capacity. They were pumping at maximum rate anyway, but actually the UAE and Saudi Arabia basically were the two that had spare capacity. So the UAE were a little bit annoyed 'cause they were like, well, we could pump more and earn more here, but we're not allowed to 'cause big brother said no. So in theory, they could pump more. Now in practice, obviously they're right in the crosshairs of this Straits of Hormuz closure. And so in practice, can they actually deliver more barrels of oil to the market? Like right now, I actually don't really know the answer to that. I'd be surprised. I'm not sure it's going to move the needle much. I mean, To back up my point, oil's just spiked to a new conflict high. This is after this news that UAE are breaking out of OPEC. So I think that kind of tells you that there's not a magic solution. All of a sudden millions more barrels of oil are going to get shipped, because I don't think that's the case. But longer term, I think it has, you know, after this crisis goes away and whatever, I think then it's an interesting change up of events.

Yeah, I was just having a quick look and it wasn't just the UAE. Scott Bessant, the Treasury Secretary, said many of the countries in the Persian Gulf have been requesting swap lines.

Yeah.

So you can imagine what's going on there.

Deals are getting done.

Deals are getting done basically. Yeah. It's like, look, we know that what we are doing is causing a disruption, but knowing full well that you are quite critical then to the ability to maneuver the price in itself whilst we know we're pushing it up.

Again, back to my point earlier about Trump and this is actually, is this playing into his hands? Is the, are we at the point behind the scenes where Trump and his team are like, actually, the longer this lasts, the better? I just wonder if they've now come around to that idea.

I guess timing's key though, right? 'Cause I was just looking at your visualizer of what the House is going to look like when the midterms roll around. And it's a sweep on the House by the looks of it. So yeah, I guess timing is quite key.

But maybe again, maybe he's just given up on that idea. Like he's lost the House. They're going to get, the Republicans will get annihilated in the midterms. I don't think he can change that now. So maybe he's just kind of written that off.

Yeah. And how do you have it? Then you just go with the strong finish to end economically. Okay, well look, let's move on and let's go on the final chapter to finish things off, which was the mega-cap tech earnings. Quite a mixed bag here. So why don't you run us through, I don't know whether you want to start with the winners or the losers first.

Well, can I just say what a night last night? Because you had the Fed meeting, you had oil spiking through $120, then you had 4 of the biggest companies on the planet. Reporting earnings within a few seconds of each other. So you had Google, Amazon, Microsoft, Meta. It was just bang, one massive drop of kind of key information. So yeah, talk about kind of news flow through yesterday evening session. But yeah, I mean, I would say, so Google, Amazon, Microsoft, Meta. I think there's two interesting angles here. Just at a top level, I would say it's Google's share price reaction versus Meta's share price reaction. Google being the best and Meta being the worst. I'd say then it's looking at Google, Amazon, Microsoft and the kind of the cloud battle. And actually Google also coming out and looking, well, having performed better than the other two. So let maybe, let's start with Google versus Meta first in terms of the best performer and the worst performer. So, as I said, Google up 6%, Meta down 6%, essentially. Although what was interesting is that they both reported operating incomes rising at 30%. So both of them, there's quite a lot of similarities from their headline numbers. What blew me, what's a great stat is Meta's revenue went up 33% and their share price dropped 6%. So their share price went down. Operating income up 30%. These numbers are just, they are just bonkers ridiculous. But yeah, their share price dropped. So, well, let's talk about Google first perhaps and why they, well, let's talk about the Google versus the Meta. Sorry. So why did one go up, one go down? I mean, Basically very similar numbers. Also both reporting big increases in CapEx spending as they both try and, you know, keep staying in this kind of superscaler race to win AI, whatever that means. So Google are upping their projects, their capital expenditure projects to about $185 billion this year. Yeah, that's $185 billion they're gonna spend in 12 months. Meta upping theirs to $135 billion. And I guess it's just the difference here, right? Is obviously it's a huge amount of money, so investors have gotta decide, well, is, are they gonna get a good return on investment for that spend? And I think this is the difference. I think now Alphabet, I mean, definitely, I mean, their numbers are insane. We'll talk about their cloud specifically in a minute, but basically their AI business model They've got the lot, right? They've got the full stack, they've got the chips, they've got the cloud, they've got Gemini, and they are, we'll talk about their cloud in a second. Their revenue on cloud is just ramping, you know, through. So it is working. It's an existing business model where the return on investment is there, right? And it's only growing. That's Alphabet. Meta. It's a bit more, well, it's a bit less clear basically. So Meta don't have any of that, right? They don't have chips, they don't have a cloud unit. And do they have a kind of a model? There's this thing that a new AI model they're calling Muse Spark. It's just not very good. It's not, you know, compared to the other, no one's ever heard of it. I bet that actually most listeners have never even heard of Muse Spark, right? Which kind of sums it up. So for them, it's about making their advertising business more efficient, making their advertising products to businesses that use it more effective. And so there it's kind of, yeah, it's just not—

Is that paying off though? I mean, you just said the revenues are up significantly. What is it that the market are expecting then? Because—

Well, think about the oil price and what we've just talked about, economic risks. If you get a recession, so is there a global recession coming? I mean, the risk of it has definitely gone up. In a recession, what kind of gets chopped first from a business's sort of spend is typically their advertising, right? So Meta are more cyclical here, whereas you could say Alphabet are a little bit more secular in that their revenue growth, a lot of it's coming through this kind of AI spend and through their cloud business, right? So I think that they're more sheltered and protected in the near-term, Alphabet, whereas Meta are more exposed. They're also spending more money in terms of their spend as a percentage of their revenue. So they're not spending more money. Meta spending $135 billion this year, Google $185 billion, but as a percentage of revenue, Meta are committing more. And then finally, I would say, don't forget about the past. Zuckerberg, who is the controlling shareholder, had an absolute shocker a few years ago when he basically bet the house and changed his entire business's name and brand to Meta off the back of this idea that they're going to spend a load of money and win the metaverse race. I mean, look, we can say right now today that was a catastrophic disaster. So his track record isn't very good on moonshot spending. So I think these factors together has resulted in this very interesting divergence off the back of the earnings numbers, which top line look very similar and yet exact opposite share price reactions.

Yeah. I, one line that I saw out of the Alphabet CEO on that side of the difference between planning and then execution, which you've quite rightly said is what they're hitting at the moment. One surprising fact I saw was Google's search queries have hit an all-time high in Q1.

Yeah.

And you remember we turned the clock back 18 months and this was like everyone was scared about the longevity of Google when AI, this is when OpenAI first broke on the scene. It was like, well, this is going to massively disrupt then the behaviors of consumers. And yet they just hit all-time highs. And then They've also had, in translated into what you were saying, the strongest quarter ever for their consumer AI plans. So the Gemini app has been a phenomenal success on that latest update they had, what, 4 or 5 months ago. So overall paid subscriptions have now reached 350 million with YouTube and Google One being the key drivers of that. So yeah.

This is the beauty of Google. I mean, right, they are just absolutely destroying it. Their revenue jumped 22% to $110 billion. But what's great about that is they are becoming less dependent on their search revenue. Their search revenue was up 19% to $60 billion, right? So it's getting towards— it's only just over 50% of their total revenue now. You know, go back 10 years, you'll have to find the numbers, but you know, we're talking 80, 90% of their revenue was search. So they are weaning themselves off that dependency. With some great diversifying products that are just going gangbusters. But at the same time, the thing they're weaning themselves off is still growing at 20, or sorry, 19%. When, yeah, a couple of years ago people had written that off. AI would destroy search and Google are in a lot of trouble was the narrative 2 years ago. And yet here we are. No one, no one, no one predicted that their search revenue would be growing at nearly 20% come 2026. It's just completely destroyed analysts' sort of bearish expectations. So yeah, it's just everything's going right at the moment for Google.

All right, to finish, before the Fed, before the tech earnings, we did have a bit of a shakeout in the equity markets in the US on Tuesday, and that was predominantly, or the catalyst was the OpenAI The Wall Street Journal reported that the company had missed its internal goal of reaching 1 billion weekly active users for ChatGPT by the end of last year. Where they landed was said to be around 900 million. The company also missed multiple monthly and annual revenue targets for 2026. A lot of shift of user base going way of, you just mentioned them, Google. But also their biggest probably rival within the Frontier Labs space, Anthropic. And the kicker here was, it was, I think it was, I don't know if it was a private closed session, but there was some photos. I think it may have been hosted by the Wall Street Journal, but the CFO, Sarah Friar, reportedly warned leadership that the company might struggle to pay its massive future computing contracts. That $600 billion infrastructure plan if revenue growth does not accelerate. Why is she saying that publicly? I've got absolutely no idea. But given the interconnectedness that we always say of this kind of ecosystem around AI, the reason why the whole equity market actually fell was chips got hit. So your Nvidias, your Broadcoms, infrastructure got hit, your Oracles, your CoreWeaves. And platforms all declined. So yeah, I thought that was, we've talked about this a lot on the show, both myself with you, with Steven, about how OpenAI were a clear early front runner. Yeah. Lots has gone on and the tide has kind of shifted, if you like. But yeah, any thoughts on what's happened with OpenAI this week?

I think it's just that they're very volatile. I mean, I'm sure I've talked about this down the years and going back to Google and why they're in such a powerful position, but like with OpenAI, it's all great until it's not great, right? The only way to, for this thing, this trick to work is if revenue growth is exponential. The problem they've got is that the competition have come in and have taken market share. And Anthropic being the key player here. And Anthropic, as you and Stephen were discussing, of course, absolutely just dominating the B2B market. So Anthropic sell business to business, whereas OpenAI is B2C, business to consumer. And just the quality of those two revenue streams are very different. So OpenAI are now struggling. You've obviously got Gemini as well as Claude from Anthropic and all the other competitors coming in. So whilst OpenAI were, you know, had the early mover advantage, the cost of this game, which is the ridiculous build-out spend on data centers to enable these, you know, your ChatGPTs to actually function. You know, if you are the end user typing away, okay, I'm gonna give ChatGPT a little bit of a kind of prompt here. And then you wait and look, we're, we want instant, we want instant answers, right? That's the, that's the consumer. We want answers now. And if you're waiting and you wait, what, what the hell's going on? Why is it taking so long? You know, that's, that's because your request has just been fired to a data center somewhere and it's actually going through that, that kind of process, right? So this is the problem. To stay in the game, it costs beyond a fortune. And it's all fine if revenue growth is exponential. It's just when you start to see vulnerabilities in that exponentiality, then you start to get concerns. And this is kind of where we're in a phase now where ChatGPT are being overtaken by Anthropic.

And so I don't know. Yeah, I saw an interesting chart and I think it was revenue and it was when you say exponential, there's like different quality grades of exponentialism, which was, I saw the OpenAI chart and it was a hockey stick and you're like, perfect. Until you see the rocket that's just gone literally vertical, which is Anthropic. It's insane. It only looks bad when referenced against Anthropic.

Yeah.

That's kind of the race you're in.

Yeah. Unfortunately, if you're being compared to Anthropic, everyone on the planet loses because that business, I don't know, it's like, it's literally unprecedented in the history of businesses.

So when you're under pressure, obviously you've gotta just keep pushing harder. And one of the things that I saw was an FT article talking about the Stargate project, which you'll remember in the last 12 months was that big project that was unveiled by the president, and OpenAI were quite center to that. Now, originally it was conceived as a $500 billion joint venture to build out and own dedicated data centers. But what's happened here is actually, I think it's the need for speed has meant that Stargate has shifted toward a more flexible approach, and OpenAI is now largely leasing capacity from third-party providers. This is then your Microsofts, which they've always had a very tight relationship, but then also Oracle, which is then one to watch, I guess, for volatility in the future ahead.

Yeah.

So they're doing that rather than constructing their own facilities, which obviously takes time. The other two factors I saw which were quite interesting was OpenAI recently halted or abandoned several major international and domestic projects, namely data centers in the UK and Norway. And then the final one was that while OpenAI executives are arguing that acting early on the compute curve gives them a structural advantage over their rivals, one of the problems is there's a clear tendency to renegotiate or walk away from projects as I've just described there. And if you are a company that's very dependent on frequent funding, that's the sort of thing that's gonna make you nervous to hand over multiple billions of dollars at this particular point in time. Plus the company's getting sued by the world's richest man on the planet. Elon Musk is in court trying to sue OpenAI and Sam Altman at the same time. So it's all going off at the moment. That red alert button, wasn't it?

Well, finally, you mentioned their strong relationship with Microsoft. That just got less strong. And I wonder whether it's desperation here because basically OpenAI, Microsoft and OpenAI, they basically loosen the terms of their, what you might call a landmark partnership from a few years back. So just signaling that obviously growing distance in that relationship, you would say. But basically this means now OpenAI are free to sell its technology more widely. So actually they've just done a deal with Amazon. So they've done a $50 billion deal with Amazon. So it comes back to funding, right? OpenAI have to keep the funding tap because they obviously, they're not profitable. So they have to keep the funding tap open by any means they can. And this, I think, looks a little bit desperate where they're now basically changing that agreement with Microsoft and going to get some more money from Amazon. What do Microsoft get? In return. So they, well, they give up exclusive rights to host the startup's models, right? So OpenAI's ChatGPT was being hosted by the Microsoft cloud, right? So they've given up that exclusivity, but that will mean, but they continue to get a share of OpenAI's revenue. And they've also secured a removal of the AGI clause. So Sam Altman in the first deal said, yeah, you can get a share of our revenues from the likes of ChatGPT, but AGI revenues in the future, you're not getting any action on that. Well, that clause has been removed, meaning that Microsoft will now get a share of revenues from any AGI action in the future. So yeah, I don't know. It's an interesting one. And maybe it's just a sign that they're desperate for funding.

Yeah. And talking of Trump and dollar swap lines and escalate to negotiate. Here you've had, oops, the mythos has got out into the world. It's going to be the end of the doomsday scenario. And then that came a few days before Elon goes to court to try and sue OpenAI. Then the Journal drops this story a few, a day or so after. Then if I was the spin doctor at Anthropic, I'd turn the screw, pick up the bat phone, call my buddy at Bloomberg and go, you know who I am, but let's just run it as sources. But our valuation for the upcoming IPO this year, we're gunning for $900 billion now. How's that? How do you like them apples? Put the phone down. Just, it's just perfect strategy from a comms perspective to juice the IPO. It's just been phenomenal on their side. Continues to be.

Yeah. It's just fascinating. Or no matter where you look, exactly. All the IPOs coming down the track with SpaceX as well. And I mean, it's just absolutely fascinating. And you've just, I dunno, it's really difficult to kind of figure all this out if you're an investor. It's just really with the US-Iran conflict and the inflation risk, And the economic risk that comes with that, coupled with this AI boom and the winners and the losers, and it's just difficult at a really top kind of macro level to have any conviction about up or down. So in the meantime, US stocks stay very high, their bond yields stay largely unaffected. But everywhere else is starting to feel the pain.

So to close, for Jerome, who I know is a big fan, big listener of the pod. Hey Jerome. What ranking, 0 to 10, would you give Jerome Powell for his time as Chairman of the Federal Reserve?

I'll give him a 6.5. Basically.

Is it the transitory mark?

It's the transitory.

Okay.

It's transitory.

I'm gonna mark him higher than that. And the reason why is purely based when Trump whipped out the paperwork when they were wearing the hard hats in the construction site on the Federal Reserve building. I thought, I thought that how he didn't react to that.

Yeah.

I thought was legendary.

And I, I've got one to finish on. Did you see in the news this week that Iran beat Trump?

No, go on.

Did you see that? It's the snooker world championships at the moment. Oh. And there just so happens to be a British snooker player called Judd Trump, who's one of the top players, top-ranked players. He's won the world championship before. Okay. Trump, he was playing the only Iranian snooker player In the world. The only Iranian professional snooker player, like literally there's only one. Oh, from Iran.

Classic.

And he beat Trump. So Iran officially beat Trump this week. There you go.

All right. There you go. Who thought we'd cover snooker on this episode? But there you go. All right. Thanks everyone for listening. Lots there to get your head around. So hopefully that was useful. We'll have our normal deals-related episode going out Monday. So thanks for listening. Have a great weekend ahead. Thanks, Piers.

See you later.