recap/history/1775430672822-lclDwnfnOGc.json

{"id":"1775430672822-lclDwnfnOGc","videoId":"lclDwnfnOGc","url":"https://www.youtube.com/watch?v=lclDwnfnOGc","title":"MacroVoices #526 Matt Barrie: Pay To PrAI","type":"youtube","topicCount":19,"segmentCount":139,"createdAt":"2026-04-05T23:11:12.823Z","uploadDate":"20260402","chunks":[{"title":"Introduction and Episode Preview","summary":"Eric Townsend previews the episode's double-header format. He introduces upcoming segments featuring Matt Barrie on artificial intelligence and Dr. Anas Alhajji on the escalating Iran conflict following President Trump's recent speech.","entries":[{"text":"Host: This is MacroVoices, the free weekly financial podcast targeting professional finance, high net-worth individuals, family offices, and other sophisticated investors. MacroVoices is all about the brightest minds in the world of finance and macroeconomics telling it like it is, bullish or bearish, no holds barred. Now, here are your hosts, Eric Townsend and Patrick Ceresna.","offset":0,"duration":28},{"text":"Eric: MacroVoices episode 526 was produced on April 2nd, 2026. I'm Eric Townsend. We've got another MacroVoices double header lined up for you, and it's going to be a doozy. President Trump gave an address Wednesday night that surprised the markets. So I'm recording earlier than usual this week on Wednesday night about an hour after President Trump's speech, in which he said that if no deal can be reached, the US plans include targeting all of Iran's civilian electric power generation plants, probably simultaneously. That's exactly the red line that Iran has previously said would cause it to retaliate by targeting desalination plants and the UAE's Barakah nuclear power station.","offset":28,"duration":53},{"text":"Eric: Now we know the president's negotiating style is to make dire threats and then walk them back once a deal is reached. So hopefully the threat of targeting civilian power generation infrastructure will come off the table soon. But overall, my take is that this conflict is heating up, not cooling down as the market was interpreting before President Trump's Wednesday evening speech. And the president was clear in that speech in saying that the US would be hitting Iran very hard over the next two weeks if no deal is reached.","offset":81,"duration":37},{"text":"Eric: We're going to kick things off with an exceptional performance by freelancer.com CEO Matt Barrie, who returns as this week's feature interview guest for a full interview on the latest developments on artificial intelligence, what they mean for private credit markets, why the AI business model is setting the stage for an eventual market dislocation on the scale of the 2000 .com bust, why paper-token inference pricing is inevitable and why it will collapse the current AI business model, and much more.","offset":118,"duration":39},{"text":"Eric: Then for this week's Iran conflict update, Energy Outlook Advisors founder and managing partner Dr. Anas Alhajji returns as this week's second feature interview guest, and it's going to run just as long as our first feature interview because there was so much to cover after the president's Wednesday night address. So this is going to be a long episode, but a very important one. So sit back, relax, and expect some really exceptional content from both of our guests.","offset":157,"duration":32}],"startTime":0},{"title":"Weekly Macro Scoreboard","summary":"Patrick Ceresna provides the weekly macro scoreboard, detailing the latest price action for the S&P 500, US Dollar Index, and key commodities, while previewing the week's economic data.","entries":[{"text":"Patrick: And I'm Patrick Ceresna with the macro scoreboard week over week as of the close of Wednesday, April 2nd, 2026. The S&P 500 index down 24 basis points, trading at 65.75. Markets put together an oversold bounce, but will rallies fail with this geopolitical backdrop? We'll take a closer look at that chart and the key technical levels to watch in the post-game segment.","offset":189,"duration":27},{"text":"Patrick: The US Dollar Index down nine basis points, trading at 99.55. The May WTI crude oil contract up 1085 basis points to 100 spot 12. The May RBOB gasoline up 439 basis points, trading at 3.09. The June gold contract up 500 basis points to 48.13. Gold's approaching its 50-day moving average, bouncing from its oversold conditions.","offset":216,"duration":48},{"text":"Patrick: The May copper contract up 162 basis points, trading at 5.65. The April uranium contract down 12 basis points, trading at 84.30. And the US 10-year treasury yield up four basis points, trading at 4.37. The key news to watch this week is Friday's jobs numbers, and next week we have the ISM services PMIs, the core PCE and CPI inflation numbers, and the FOMC meeting minutes.","offset":264,"duration":28},{"text":"Patrick: This week's feature interview guest is freelancer.com founder Matt Barrie. Eric and Matt discuss the sustainability of the AI business model, why rising risks and weak unit economics could lead to a shakeout similar to the .com era, and how AI's impact on software and private credit could create broader market stress. And stay tuned for a special follow-up with Dr. Anas Alhajji where we break down the escalating Iran conflict and its implications on global energy markets. Eric's interview with Matt Barrie is coming up as MacroVoices continues right here at MacroVoices.com.","offset":292,"duration":48}],"startTime":189},{"title":"OpenAI's $122 Billion Fundraising Round","summary":"Matt Barrie breaks down OpenAI's massive $122 billion bridge round ahead of its anticipated IPO. He highlights the extreme reliance on vendor financing from tech giants like Amazon and Nvidia over actual cash.","entries":[{"text":"Eric: And now with this week's special guest, here is your host, Eric Townsend. Joining me now is freelancer.com founder Matt Barrie. As usual, Matt has written an excellent paper about AI. This one is called \"Pay to PrAI,\" that's pay to P-R-A-I. You can find that linked in your research roundup email. If you don't have a research roundup email, it means you're not yet registered at MacroVoices.com. Just go to our homepage, MacroVoices.com, look for the red button above Matt's picture on the homepage that says \"Looking for the downloads.\"","offset":340,"duration":43},{"text":"Eric: Matt, before we even get into all of what's going on in AI, we've got some news just as we're recording this on Tuesday evening US time. There's some recent news just in the last day or so, which is, you know, a lot of companies before they IPO, they'll do a little bridge round, maybe 10-20 million bucks just to cover some expenses until they get to their actual IPO. OpenAI is planning an IPO. They did a little bridge round, 122 billion with a B dollars. All-time record for a private fundraising round. And what is that, something like 100 times bigger than any private fundraising round has ever occurred like before 2025. Why did they need 122 billion just to bridge them from now to their IPO, which is expected in less than a year?","offset":383,"duration":45},{"text":"Matt: Thanks for having me. It truly is stupendous. I mean, before 2025, the largest sort of private venture rounds were in the single-digit billions. The headline number is 122 billion raised on a 730 billion pre-money valuation. When you get down to the actual segmentation of kind of what's going on, it seems that it's only about 25 billion dollars worth of cash. It seems to be more of a vendor financing than an actual straight cash injection.","offset":428,"duration":33},{"text":"Matt: You've got Amazon, Nvidia, and SoftBank primarily putting the money, or the in-kind, in. Amazon's putting in 50 billion, but that's contingent on OpenAI spending 100 billion, I think over the next eight years on their compute, which I'm sure won't make Microsoft happy, and it seems like a good deal, at least on paper, for Amazon. So there's 15 billion going in up front and 35 billion is furthermore contingent on the company either going public by 2028 or achieving artificial general intelligence. I'm not sure how they're going to really define that. I think the underlying definition is a panel of experts will make a decision, yes or no, which is a bit strange for a financial decision.","offset":461,"duration":33},{"text":"Matt: There's 15 billion down up front, 35 billion contingent on going public or AGI for a 100 billion commitment the other way around, so it's a bit like a procurement round. SoftBank loves doing this sort of, I like to call them Russian stands, where they kind of lead rounds and mark up valuations to the moon. We saw it with WeWork, we're seeing it again with Enron, I mean OpenAI. They've already put in about 40 billion, they're putting another 30 billion in. But to raise that money, they don't actually have the cash on the balance sheet, so they've taken a 12-month bridge loan for 40 billion and they're tranching in 10 billion at a time over the course of the year for a total of 30 billion.","offset":494,"duration":37},{"text":"Matt: And obviously that's kind of just getting them through to the IPO so they've got a liquidity event. And then Nvidia is putting in in-kind, as they tend to be doing in all these sort of circular economy sort of deals in the AI space, where GPUs and infrastructure will be provided to the tune of 30 billion into the round. So it's about a 25 billion dollar round of cash sort of up front, 10 from SoftBank, 15 from Amazon. We'll see if the rest of the money comes in, but the rest is in-kind.","offset":531,"duration":27},{"text":"Matt: So it seems from looking at this, and if you look at the compute numbers, I mean they truly are astronomical that are being contributed in the form of, you know, either Nvidia credits, I think there's like three gigawatts of inference and two gigawatts of training capacity as part of this investment. You know, that's sort of the power that gets drawn by a small country. So it seems to be what they're trying to do is scale up the spend on compute so much that they can find a way to bring the unit economics down because that's really the key problem in the space, that nobody in the AI compute space is making any money other than really Nvidia who does about 160 billion of revenue and 100 billion of earnings, and then TSMC who provide the chips to Nvidia, but the rest of the space is actually negative on using the product in terms of the unit economics. So the more you use the product, the more you lose the money. So I think they're trying to make it up on volume.","offset":558,"duration":49}],"startTime":340},{"title":"The Unsustainable Economics of Consumer AI","summary":"Matt details the massive cash burn required to train and run foundational AI models. He explains how computing costs drastically outpace Moore's Law, rendering popular consumer subscription tiers highly unprofitable.","entries":[{"text":"Eric: Matt, I can't imagine any responsible business executive signing off on 122 billion dollar deal unless the underlying business model was rock-solid and it didn't have any major risks in it. Just happens that you wrote this missive over the last couple of weeks called \"Pay to PrAI.\" I don't think that's actually the conclusion that you reached. Tell us about the economics of the, I'll call it the consumer AI business model, offering AI through chat to people like me who sign up for a Max subscription on Claude or a Pro subscription on OpenAI. How much money are they making or losing on that?","offset":607,"duration":37},{"text":"Matt: The AI industry is consuming an absolute bonfire of money. It's about 600 billion dollars a year that's being spent by the hyperscalers on CapEx. It's reaching a kind of a point now which is kind of incredible where, you know, the CapEx is higher than their internal free cash flow. You know, the fundamental business model that's being pushed in the consumer market and the software development market up until now has really been a venture capital style subsidized model where you pay, you either use the free product and then hopefully upgrade to the 20 dollar a month product, or you use the 20 dollar a month product if you're a consumer or a 200 dollar a month product if you're a power user or if you're starting to do programming.","offset":644,"duration":38},{"text":"Matt: But the problem with these models are that they take an incredible amount of money to train, and I think we've talked about that in previous episodes where you've got training runs North of 100 million dollars a run, you know approaching half a billion dollars a training run. That requires a huge amount of data center build-out, a whole lot, an incredible stupendous amount of data needs to go into these models. And so, you know, because the cheap data that's scraped off the internet for free is basically sort of, you know, drilled out to an extent, they have to do licensing deals to get access to that data. Sometimes they do dodgy things like Anthropic, you know, scraped a whole bunch of books and scanned them in and then they got caught and they had to pay the biggest, I think, fine in copyright history, I think it was about 1.5 billion dollars for the illegal scanning of all that data, etc.","offset":682,"duration":42},{"text":"Matt: So it's incredibly expensive to train. But the fundamental problem is that when you actually use the inference, you basically put queries into GPT or queries into Claude and you run them, those queries are loss-making. You can't make it up on volume under the current models. While we do see, you know, the underlying hardware is on sort of a Moore's Law sort of trend, and Moore's Law for those of you that don't know what it is is, you know, every 18 months to two years or so, effectively the technology that goes into chips allows semiconductors to be produced with finer and finer feature sizes, which effectively allows the compute capability in terms of processing power to effectively double or the cost to half every 18 months.","offset":724,"duration":44},{"text":"Matt: While you are riding on that sort of silicon Moore's Law trend, because the models are in such a brutally competitive environment where there's literally zero lock-in, there's almost zero switching costs, so I can, if one day ChatGPT 5 comes out I'll switch to that, but then Claude 4.6 opus comes out and that's better, I'll switch to that, and so there's nothing stopping me overnight really just changing the models. Because of the competitive environment and so forth, what the situation basically is that each generation of model is being rushed out to kind of get the top of the scoreboard so that all the customers flock to that.","offset":768,"duration":35},{"text":"Matt: And as a result of that, the actual amount of inference or tokens burned, if you will, per useful query with each generation of model, while the underlying compute is getting cheaper and cheaper in terms of what you can get done on the chips, the amount of inference you have to burn for a useful query is actually going up quite dramatically. And so you're actually not seeing a reduction in the cost of inference, you're seeing an increase. And that's before you consider the issue with energy and constrained build capacity for building data centers and power gear and all the other things.","offset":803,"duration":36},{"text":"Matt: And so, you know, a few people have done various models of how much it costs for GPT to deliver their 20 dollar plans, and even if you ask Claude itself, you type a query into Claude saying how much is the underlying compute costs for on a 20 dollar plan with Claude, it will tell you, you know, 15 to 20 dollars, maybe 18 dollars, maybe a bit more. So effectively there's no money being made on these 20 dollar plans and when you're a power user you can burn up to several hundred dollars on a 20 dollar plan. And it gets even worse on these 200 dollar plans which are used by programmers now, because the nature of programming is you're streaming tokens almost forever, and you know, the average user on a 200 dollar plan can burn, you know, many thousands of dollars of underlying compute, and in fact there's a, there's a leaderboard called VibeRank where people compete to see how much underlying compute they can burn on a 200 dollar plan. The leading guy on the leaderboard has burnt 51,000 US dollars in a single month on a 200 dollar plan.","offset":839,"duration":60},{"text":"Matt: So the issue is that as these models get more and more competitive and new versions come out, you know, you're getting an exponential increase in the amount of inference you've got to burn, and that is meaning that these models are not getting cheaper and you're not making up on unit economics. And so it kind of feels like this funding round, especially with the amount of vendor financing that's packed into it, it feels like this is an attempt to really try and scale up the underlying infrastructure and GPU capability so that potentially you know, some sort of threshold can be crossed in the underlying economics so inference could be profitable. But there's one big problem with all that, which is is the demand there? And and that's where we are.","offset":899,"duration":44}],"startTime":607},{"title":"AI Valuations and Dot-Com Bubble Parallels","summary":"Eric and Matt compare the current AI funding frenzy to the late 90s dot-com boom. They debate whether companies like OpenAI can justify trillion-dollar valuations given the lack of sustainable competitive moats against open-source alternatives.","entries":[{"text":"Eric: Matt, you said it's 122 billion dollar capital raise on a pre-money of 730, so I get 852 billion is the current enterprise value. Assuming an up round, which is what everybody assumes, we're going to go just let's call it an even trillion for the IPO because hey what's, you know, 100 billion here, 100 billion there, you know, eventually you're talking about real money. But holy cow, let's say that OpenAI goes ahead and IPOs sometime this year for 1 trillion dollars. Are you going to be long, short, or flat, and why?","offset":943,"duration":37},{"text":"Matt: It really does feel in the space that we're really in the moment where a supernova is starting to explode and we're probably going to end up with a giant black hole like we did in the .com boom the first time around. I mean a trillion dollar valuation at IPO is absolutely gigantic. You know, they've raised 122 billion here in this round. They were mooting that the IPO earlier in the year was going to be a 60 billion dollar raise on about a trillion, but you could probably imagine that that number has possibly scaled up a little bit or they want to keep it tight obviously because once it gets to public markets you don't want to have too much stock unload onto the market.","offset":980,"duration":36},{"text":"Matt: But a bigger problem for them is the fact that you've got Elon Musk in the wings who's not just suing OpenAI because it turns out you can't IPO a charity and they've converted OpenAI to a for-profit model and there's a lot of complications around that, but he's also IPOing his SpaceX for I think it's 1.75 trillion valuation. So there's a big possibility that a lot of the heat's going to be taken out of the market when he does his 75 billion on a 1.75 trillion dollar raise, and it looks like he's going to beat OpenAI to to going public because if you kind of look at the news and you know what's all happening with the ETFs and the allocations and so forth, he's a lot further along.","offset":1016,"duration":40},{"text":"Matt: I don't know, in the olden days, you know, if you were to invest in Amazon when it went public or Microsoft when it went public, you know, you had the ability to make a lot of money. I don't know how much is left on the table for the general public when the company's being valued at a trillion dollars when it goes public, you know, what are you going to do? Get to two trillion, 10 trillion? I think Nvidia is what, four or four and a half trillion valuation, and they're the only ones making money in the space.","offset":1056,"duration":26},{"text":"Eric: Matt, the parallels between this and the late 1990s .com boom before the 2000 .com bust are just striking to me. And it occurs to me before I go on that we probably have listeners that weren't even born when that happened. So for anyone who's not familiar with what happened there, Wall Street became absolutely obsessed with the idea that the internet is going to be a really big deal. And the thing that's really important to understand about this is they got that call exactly right. The internet, the public internet was going to change the world we lived in in ways beyond what anyone could even conceive and they got the call right that it was a really big deal.","offset":1082,"duration":44},{"text":"Eric: And I think they're getting the call right again that AI is a really big deal, maybe as big or bigger than the public internet. But the thing is, even though they got the call right, they started throwing money at dumb ideas without thinking and it just turned into a complete frenzy. It seems like that's happening again here, but in the late 90s, it was every tiny little company that had .com in its name and it didn't really matter whether they had a business model. It's different here. It's the big players, which frankly have a business model. But as you very eloquently explain in this excellent piece I recommend everyone read called \"Pay to PrAI,\" that business model isn't viable for the reasons that you just described, or it's not profitable or not likely to be sustainable.","offset":1126,"duration":50},{"text":"Eric: How should we think about this? Is it inevitable that a .com bust like we had in 2000 is coming for AI, and does it have the same dimensions as the one in 2000? Is it bigger, smaller, worse, better? What do you think?","offset":1176,"duration":16},{"text":"Matt: So if you think about the .com boom, and I was there actually in Silicon Valley in '97, '98, '99, 2000, I saw it all. There was a hypothesis the internet was going to be a big thing and it turned out to be a enormous thing in terms of the benefits to humanity and society and will continue to grow in terms of its applications. And AI is the same thing. AI is going to be absolutely transformative for humanity in terms of what it can do. At that time, you know, a company whose valuation went through the roof was Cisco, and their tagline was \"We network networks.\" We talked about it I think last MacroVoices. Every time you plugged in a bit of the internet, you needed to have a router to connect up the network and Cisco equipment was going to be everywhere and why wouldn't it be the most valuable company in the world, right?","offset":1192,"duration":42},{"text":"Matt: And the same as with OpenAI in a to an extent, you think okay, well they've got the best AI models, AI is going to be everywhere, why wouldn't OpenAI be the most valuable company in the world? But then at the same time, you also had AT&T, and up until about 1996 AT&T had a 60% market share. It basically built the network using Cisco equipment to connect up the world. You had three players in the market and I think it was mentioned at the time by an analyst that they had margins that would make drug dealers blush.","offset":1234,"duration":25},{"text":"Matt: But the problem was that in 1996 the Telecommunications Act, Regulation Act came in and deregulated the market. And you started having fourth entrants and fifth entrants and so forth and and people buying and selling capacity to each other etc. and when you started having the fourth entrant come in and competing on cost, the unit economics fell apart. And if you think about the AI compute space, well Amazon had it pretty good up until the AI boom, I think its its CapEx as a percentage of earnings was down to about 6% at one point.","offset":1259,"duration":30},{"text":"Matt: Now these hyperscalers are spending, or the cloud computing companies are spending, 60% of earnings on CapEx, over 100% of earnings on CapEx. They're spending 600 billion dollars a year at the moment in terms of a run rate in CapEx and it's hypothesized that by 2030 it's going to be 5.2 trillion dollars of CapEx, right? Now when you just had Amazon, you know, really dominating in terms of of cloud, you know, it was sitting pretty and I think it had about 60% market share.","offset":1289,"duration":26},{"text":"Matt: Now you've got Microsoft who's taken over and they're market leader and so forth. Now you've got entrants such as Oracle that's competing on price and you've got CoreWeave and the Neo clouds and the unit economics are starting to look a little bit shaky. And then you had a big bust in the .com bust where, you know, all these companies had money being thrown at them and you're seeing that right now. Any company that kind of has AI in its business plan is getting these stupid, you know, seed rounds in the hundreds of millions of dollars. I remember when seed rounds were hundreds of thousands of dollars, now you have hundreds of millions of dollars.","offset":1315,"duration":36},{"text":"Matt: And you know through the early 2000s you had a real trough in the technology space where where everything blew up and was somewhat uninvestable for a period of time. But through that period of time, you had companies like Google and Amazon continue to grow and double down and Microsoft and so forth and ultimately today became very very big companies. So I think we're seeing this on steroids right now with AI CapEx and compute and the OpenAI funding rounds and so forth. And you know it's unquestionable AI is going to be completely enduring and just absolutely game-changing for society. But I think the kind of circle jerk of money that's sloshing around between a very very small number of companies at stupidly high valuations and at a scale that is just stupendous is going to potentially end up in a big bust.","offset":1351,"duration":51},{"text":"Eric: Matt, I want to focus on that because a lot of people are going to mishear what you just said and they're going to say Matt Barrie said imminently tomorrow there's about to be a great big bust in all of the AI stocks. That's not what you said and it really rings home for me because the reason that my partners and I sold our software company in the summer of 1998 is because we knew it was a bubble. And my plan at the time was to take all of the proceeds and short the NASDAQ because I knew it was a bubble.","offset":1402,"duration":33},{"text":"Eric: Fortunately I got talked out of that, but if I hadn't, I would have lost everything because even though we were right, basically the NASDAQ doubled between '98 and 2000 before it crashed exactly like I predicted it was going to crash. What do we do here? I mean, do you go long, do you go short? And I'll say I'll ask the question this way: OpenAI and Anthropic are both expected to IPO probably in 2026. When they do, you know, is it time to go long or is it time to go short? Because you know I could make either argument. It seems to me eventually you want to be short, but how do you time that?","offset":1435,"duration":37},{"text":"Matt: That's the trillion dollar question, right? You've got SpaceX going public which obviously has Grok and xAI within it. You've got Anthropic going public and you've got OpenAI going public and that's why I think it feels a bit like a supernova, so we'll end up with a big bang in one way or or another. I mean, you know, I think the issue is going to be, I mean that some of these valuations like the OpenAI valuation is kind of predicated on the fact that it's going to capture an enormous amount of value out of the world in order to justify these valuations, right? You know, these funding rounds are so large and the valuations are so high that you need to pitch a revenue line that matches them.","offset":1472,"duration":36},{"text":"Matt: And you know at the moment OpenAI is doing about 2 billion a month in revenue, I think it's at about a run rate of about 25 billion, I wish they'd stop using the word run rate because I think you've actually got to recur once before you can call something annual recurring revenue. So it's doing about 2 billion a month of revenue, losing you know 14 billion this year, I think the burn rate that came out today was is expected to lose 70 million dollars a day this year scaling to 156 million a day lost next year. But in order to to, you know, cross the valley of death and get to the promised land, they've got to show a business model that makes sense.","offset":1508,"duration":34},{"text":"Matt: And I think what their business model is going to is twofold. One is that they're going to take a substantial amount of white-collar jobs away from humans, you know some substantial percentage of jobs in the world are going to go to AI and that's kind of the top-line revenue number. And then the in terms of generating earnings out the other side, the justification is that the infrastructure that's being contributed as part of this 122 billion dollar round is at such a scale that only OpenAI will have the unit economics that will make the inference profitable. So in a combination of taking everyone's job and having the only infrastructure that can run this stuff profitably, I think that's the argument for justifying these sort of trillion dollar valuations.","offset":1542,"duration":42},{"text":"Matt: Now I don't think OpenAI is going to capture, for example hypothetically, you know the value in the AI-powered drug discovery market anymore than AT&T captured the value of the iPhone, right, you know? The underlying technology phenomenal, but as Ilya Sutskever said himself, who was one of the founders of OpenAI, that you can just read 40 academic papers and 95% of what's out there in AI is is published in in the public domain. And that's why every couple of weeks there's a kind of a new foundational model that's top of the leaderboard and every once in a while there's a team of people that you've never heard of, that 160 engineers in a room in Hangzhou like like DeepSeek that comes out of nowhere and suddenly leads in the leaderboard because it turns out that there's no sustainable competitive advantage in the underlying foundational models and what you do need though is access to data.","offset":1584,"duration":44},{"text":"Matt: Those data sets are now kind of out there in public or in the case of the Chinese they probably don't care too much about copyright so they'll always have a sustainable competitive advantage. And I think there's incredible and tremendous opportunity in AI, but it's probably not going to be in these models that that haven't really figured out what the business model is.","offset":1628,"duration":23}],"startTime":943},{"title":"Apple's Edge AI Strategy vs. Cloud Compute","summary":"The conversation shifts to Apple's potential to capitalize on a future AI market correction. Matt suggests Apple can avoid the data center capital expenditure bonfire by pushing AI workloads directly to local edge devices.","entries":[{"text":"Eric: Matt, I think there is a very distinct difference between this scenario and the late 90s to 2000 .com story, and that is this. Let's say that we don't know whether it's '97, '98, or '99 right now, but we know that March of 2000 is coming at some point. There's going to be a washout in this commercial AI space where they just realize that the business model wasn't sustainable. Well what happened in the .com bust is we went really a good solid couple of years of kind of a technology recession where there wasn't a whole lot of progress on the public internet because we had to basically shake that malinvestment out of the system, get back to efficient capital allocation, and then you know after 2002, 2003 things started to really take off again.","offset":1651,"duration":51},{"text":"Eric: What if the US government steps in and says, now wait a minute, the military applications of AI create an existential threat to the country if we don't stay ahead of this? We can't tolerate a 2000 to 2003 pause. You must continue, you must do it under US government funding, but we're not going to fund the giving away stuff for free to Max subscribers on Claude. We're going to take it over and you know it's just for the military now. We're not going to have consumer AI anymore. Is that a realistic scenario, or do we need consumer AI in order to train the models in order for the military to get the benefit, you know, what would happen in that scenario where the military says no, we we can't allow a stock market crash to slow down progress on the military applications of AI?","offset":1702,"duration":49},{"text":"Matt: So I don't think it's realistic that we're not going to have consumer AI. I mean the Chinese are open sourcing their models and in fact they've got their strategy to open source both the software and the hardware. You know, and there's been previous leaks of Meta's Llama and and so forth. I think the interesting thing here is that it won't be the government stepping in. I think there's one big company that's kind of been sitting in the wings that has a God-awful AI product and it's in everyone's pockets and that's Apple and the iPhone.","offset":1751,"duration":25},{"text":"Matt: I think in our last chat we talked about how goddamn awful Siri is for supposedly being your original chatbot AI assistant and we hypothesized that it will continue to be awful in the next year and here we are and it it continues to be pretty bad. But they've kind of avoided this whole getting sucked into this whole CapEx bonfire and they've really got two levers they can pull. One is is they're sitting on this cash and they've kind of they can sit back and watch and and as with the .com bust, there was a lot of infrastructure that went through to the second and third owners where you know maybe the first owner that builds out the optical fiber network goes bust and the second owner comes in and tries to make it break even after after purchasing it at a at a cheaper price and washing out the underlying sunk costs, and then the third owner comes in and makes some money out of it.","offset":1776,"duration":46},{"text":"Matt: And I think you might have that situation in the AI compute space where if there is a problem with some of these large foundational model companies, there might be a second or third owner and Apple would be ideal for that. At the same time, what they've been doing is they've been putting in some AI silicon into their products and a lot of this compute that's currently being done in a data centers, you know, by OpenAI, by Anthropic, etc. is going to go to the edge and it's going to go to the edge for a variety of different reasons. One is that you don't want Sam Altman training on your data, and I think we've talked about before that I think an emperor has no clothes moment is heading into into SaaS where you know I think large enterprises are starting starting to think to themselves, you know what, I don't want my data in Google Drive and Gmail, you know it might get trained on.","offset":1822,"duration":44},{"text":"Matt: And you're certainly seeing all the major SaaS companies quietly flicking on a switch in the settings without telling you where they're saying by default now we can train on your data potentially, and then you flick it off but by then it's too late and all your data has been sucked down. So a lot of that compute is going to go the edge, it's going to go to on-prem, you know in the enterprise on compute devices so it doesn't go into the cloud. It's going to be on your phone, it's going to be on your MacBook, it's going to be on your Mac Studio or whatever it may be for privacy, for confidentiality, for latency reasons and the fact that there's some models now that can actually fit on that silicon.","offset":1866,"duration":34},{"text":"Matt: And you know while it might be you know one or two generations away, you know at some point I think real soon a lot of that load and a lot of that compute that's going to AWS data centers and Azure data centers to run Anthropic and to run OpenAI is going to go to the edge. It's going to be on Apple's devices, it's going to be on Apple's laptops, it's going to be on-prem, at the same time Apple's cashed up and might end up being the second or third owner of some of these companies.","offset":1900,"duration":26}],"startTime":1651},{"title":"The Inevitability of Per-Token AI Pricing","summary":"Matt predicts AI platforms will eventually be forced to abandon subscription plans in favor of pay-per-token pricing. He argues this dynamic will turn software development into an expensive, unpredictable financial slot machine.","entries":[{"text":"Eric: Matt, the title of your missive that you just written, \"Pay to PrAI\" or \"Pay to PrAI\" P-R-A-I is actually a reference to the inevitability in your prediction of something called paper-token monetization. What does that mean, how does that relevant, explain what that's about?","offset":1926,"duration":19},{"text":"Matt: Well, the fundamental business model of Silicon Valley venture capitalists is to try and win markets by financing companies with astronomical amounts of money such that that money gets spent on marketing and subsidization of the product to such a scale that nobody can compete with these Silicon Valley invested companies or unicorns. And so total nuclear war is launched on a market, so you can think maybe Uber and free rides in China or whatever it may be or you know DoorDash delivering noodles to people in Indonesia or you know Grab or whatever whatever it may be.","offset":1945,"duration":35},{"text":"Matt: So what we have here in the AI space, the subsidization model is that you have a free product being GPT or what have you, and then you've got a 20 dollar a month product which I've said earlier probably costs them 20 dollars to serve, and a 200 dollar a month product which probably costs them 2000 dollars to serve. You just can't continue forever. At some point, so the so the question is going to be can we get the unit economics down to do inference to such a point that these models are profitable before these companies run out of funding?","offset":1980,"duration":30},{"text":"Matt: Now the problem is that as we've moved into for example software development which consumes a never-ending amount of tokens to write code because every company in the world is powered by software now, you know there's this huge token burn that's happening in these 200 dollar plans. And if you were to try and make some sort of reasonable software-like margin, like 80% etc., you would have to price these plans not at 200 dollars a month but maybe a thousand or two thousand dollars a month or higher.","offset":2010,"duration":30},{"text":"Matt: At some point the money is going to run out and I think they had a near-death experience in the last couple of weeks. Anyone who and I know you're constantly using GPT and Claude etc. and you probably noticed the same thing, you know in the last couple of weeks there seems to have been a bit of a panic from these companies where you know you log into your 200 dollar plan, you type a couple of queries and then you run out of credit. You've always known that these companies aren't making money on the inference because instead of saying put your credit card in and top up your credit, it puts you in the naughty corner for seven hours or longer.","offset":2040,"duration":31},{"text":"Matt: The inevitable destination for these subscription models which are basically massively subsidized Silicon Valley financed and increasingly debt-financed business models is that they're going to have to move to a per-token pricing. Now and that is going to cost a lot of money. I think there was some comments in the last couple of weeks on Reddit where someone who's on a 200 dollar plan ran out of credit pretty quickly, he had to get it done he had to get something done you know that night, and so he moved to the API pricing which is in the programming interface and that was costing 200 dollars an hour instead of 200 dollars a month.","offset":2071,"duration":34},{"text":"Matt: And he thought gee, I've just you know used a couple hours used 5, 600 dollars. The problem with that sort of pricing has several dimensions. The first is that already at 20 dollars a month for a plan which is 240 US dollars a year, that already prices the product out from over half the world's population. I mean the median global income in the world is about 2.5 thousand dollars a year so at 240 dollars a year you're already 10% of the pre-tax income for half the people on the planet, right? So you're already quite expensive.","offset":2105,"duration":33},{"text":"Matt: The second point problem here is that you know in these programming models which which realistically to make any sort of margin need to be priced in the thousands of dollars a month or maybe 10,000 dollars a month, you have a bit of a problem in the fact that you know these models do hallucinate. And when they do hallucinate and throw errors, they're very different from human. When you use hire a freelancer for example on my website freelancer.com, you know you put in some money, okay I'm going to pay you 200 dollars to build a website for me and you don't release that milestone until the job is done.","offset":2138,"duration":33},{"text":"Matt: And if the human can't figure out a problem, you know they'll they'll ask their friends, they'll try and find a more senior engineer for advice, they'll get on a forums and ask other people how to solve problems, they'll browse the internet, they'll they'll kind of hill-climb their way out of solving problems. And look, they may they may ultimately not get there and you might get frustrated with them and want to find another developer to do the job, but ultimately their failure modes are very different from AI. With with AI, sometimes when it hallucinates, it can do wildly crazy things.","offset":2171,"duration":28},{"text":"Matt: About three weeks ago, I had a problem with my with my VPN on my computer, I was using Claude to kind of help me through figuring out how to get it fixed. I'm a software developer by background, I've got electrical engineering degree from Stanford, you know I I do know how to program quite well, but I kind of veered into PowerShell commands in Windows 11 which I don't know very well and I was blindly pasting in Claude and it made me delete my entire networking stack.","offset":2199,"duration":26},{"text":"Matt: And so you have these very crazy failure modes with AI where you either go in loops or it goes away and thinking and and you know burns you know 10x the inference you know try you know these reasoning chains trying to figure out what's going on, or it just has these crazy suggestions and and it will look at you in the eye with the the eyes of a sociopath on a first date in some regards trying to you know gaslight you into thinking that its answer is is true when it's clearly not the case, it's just hallucinated something or rather.","offset":2225,"duration":32},{"text":"Matt: And the issue here is that in a paper-token pricing model, it really turns software development into a slot machine in that if the if ultimately you know you're writing your app and you you kind of need to get Claude to do something or rather, you're really pulling the slot machine handle, you don't know how much it's going to cost you by the time the tokens are all burned and you don't know if you're actually going to get a solution at the end of the day. And so, you know I guess only in Silicon Valley could they turn software development into degenerate gambling because that's that's kind of where you end up.","offset":2257,"duration":33},{"text":"Matt: And the frustrating thing I think will happen is when you're on the 200 dollar plan you've got a certain amount of capacity and maybe it tells you to time out or what have you, you kind of know, you're capped at 200 dollars, right? It's very frustrating what it runs out of credit you gotta find alternative ways potentially of doing things, but you kind of know what you're what you're up for. When you're pulling the handle under the paper-token model, it could be 50 dollars per spin, you may go in a circle, you may go in a circle 10 times, you know there's all these examples of people get over X of of radical crazy things that that that Claude does to your codebase.","offset":2290,"duration":36},{"text":"Matt: And I think people are going to get very very frustrated if they have to kind of put a coin in the machine, pull the handle every time and they get a non-deterministic outcome of whether they're moving forward in a hill-climbing sense to to their final solution and their final app or their final bit of software being developed or whether they're going in circles you know round and round and round round again. And I think people are going to get frustrated, they're going to go try and find open source models, they're going to try and find alternative ways to get things done and I and I think at that point, that's where the bubble starts to pop, it's really the problem.","offset":2326,"duration":40},{"text":"Matt: And while the hallucinations are reducing as part of you know each new generation in terms of the engineering of the infrastructure around the foundational models to reduce those hallucinations, what is actually happening is the token spend is going up exponentially because you're doing more and more complex things, so you have a much bigger surface area in which you could generate an error. So what I mean by that is, you know, while the probability of a failure pulling the pulling the handle on the slot machine is getting reduced with the you know the engineering from each subsequent generation of model, the number of handle pulls you need to make is is going up exponentially and so you've got a multiplicative effect in terms of potentially the impact of errors.","offset":2366,"duration":30},{"text":"Matt: I just think it's going to, the question that basically needs to be solved now with this 122 billion dollar fundraise is can you get the cost of compute way down to get this whole model profitable and will the market tolerate software development as a slot machine.","offset":2396,"duration":6}],"startTime":1926},{"title":"AI Dependence and Token Budget Constraints","summary":"Eric shares his anxiety over recent Claude outages and professional dependence on AI. Matt explains how AI companies are quietly restricting user token budgets to survive their aggressive cash burn rates.","entries":[{"text":"Eric: A lot has been said already by lots of people about the incredible rate of progress and how quickly AI itself is getting smarter. What I don't think has been discussed enough and I'd like to get your comment on is the rate to which professionals are becoming dependent on it. I think more than you know this it's more addictive than cocaine. I remember our first interview on AI when ChatGPT had just come out and I remember thinking to myself, boy Matt's really into this stuff, to me it's a novelty but I don't think I'd ever pay 20 bucks a month for it. I mean it's just it passed a Turing test, big deal but I don't think I'd ever buy a subscription.","offset":2402,"duration":46},{"text":"Eric: I'll tell you Matt, in the last several weeks, there's been a lot of stress in my life because of this war and family that are affected by the war and waking up double digit percentages down on a percentage basis on my net worth because of something that happened in the market overnight. Okay look, I'm a big boy I've been through that stuff before, I've been through the 2008 crisis, it's not that big of a deal. But Claude 4.6 was going offline and the server wasn't available and I was freaking the F out. I couldn't handle it. I was losing, and don't you dare insult me by suggesting that I go back to ChatGPT 5.4. I don't drink Pabst Blue Ribbon and I don't do ChatGPT, okay? I mean I've gotten to the point where I can't live without Claude. This is this seems like a risk.","offset":2448,"duration":55},{"text":"Matt: When Claude went down, it's quite interesting actually because the first time it had a bit of an outage was when the data centers got blown up in in Dubai and you kind of have to wonder to yourself what potentially was being run in those data centers in the Middle East that caused the outage of Claude. And I do think those companies did have a bit of a near-death experience in the last couple of weeks. Anyone who and I know you're constantly using GPT and Claude etc. and you probably noticed the same thing, you know in the last couple of weeks there seems to have been a bit of a panic from these companies where you know Sam Altman killed Sora which was this hyped-up video modality model where you could generate you know clips or you know the whole point was you were supposed to be able to type in a prompt and get a movie out the other side.","offset":2503,"duration":31},{"text":"Matt: In fact Disney paid a billion dollars to OpenAI for use of the technology and only found out half an hour before a meeting that the whole thing was going to be canceled. At the same time Sam Altman killed instant check-out, he's working on some sort of erotic model as well which is a bit strange but I think it's probably one of the big markets is pornography and he thought maybe we could make some money there. He killed that as well and at the same time Claude had these big changes in terms of how they did the plans and they were giving out you know extra tokens in off-peak but in peak they'll kind of cutting you back etc. and then ultimately it seems that they've they've cut back a lot of lot of the token budget you get in these models and on the path to this sort of paper-PrAI paper-token sort of business model.","offset":2534,"duration":40},{"text":"Matt: So I do think they all had a bit of a near-death experience and that near-death experience obviously is what I was feeling or what you've been feeling when you kind of see these models you start using them and going, gee is my access going to start getting restricted, you know will I have to pay a lot more money, will I have to add a zero to my to my monthly subscription, what what's going on. And then of course you've got this financing round which is less cash and more infrastructure and perhaps it's a way to kind of help the company kind of limp towards IPO so there's liquidity event so the you know original investor can kind of make a bit of a return.","offset":2574,"duration":28},{"text":"Matt: It turns out that you know this whole, it's pretty funny that this whole AI compute space is predicated on 600 billion dollars a year of CapEx which is incredibly energy-intensive when at the same time you're bombing the you know an area where 48% of the world's energy is and you know you're relying on energy being cheap in order to have this AI AI boom work.","offset":2602,"duration":24}],"startTime":2402},{"title":"AI's Threat to Private Credit Markets","summary":"Matt explains how massive debt financing for data centers threatens to destabilize private credit markets, particularly because AI pitches aim to disrupt the very SaaS companies underpinning these debt portfolios.","entries":[{"text":"Eric: Let's move on to private credit. There's a lot that's been said already as a lot of private credit funds are gating investors that this is AI driven or it's related to Claude Code specifically creating fears that software companies would no longer be profitable. And I have a hard time with this one because this sounds like the claim I remember hearing in the 1970s when supposedly the introduction of the Hewlett Packard electronic calculator was supposedly going to put every accountant out of business and create vast unemployment of bookkeepers and so on and so forth and it was the exact opposite. It created more productivity. It seems to me that Claude Code just gave the software industry the biggest productivity boosting tool that they've ever had, and that's the reason that private credit is blowing up? Is that really right? Am I missing something?","offset":2626,"duration":54},{"text":"Matt: Well, I mean there's a few things going on. First of all, the these funding rounds are getting too big for equity, so as we saw with 122 billion and all the infrastructure, the vendor financing in there. And so you've increasingly these companies are turning to debt, even Meta had to go and get 30 billion from Blue Owl not so long ago in order to fund data centers because it's [44:59]","offset":2680,"duration":20},{"text":"Host: ...not do it off the balance sheet anymore and the numbers are starting to get too big for equity raising, so they're borrowing the money. And not just are we seeing record equity rounds, we're seeing record debt rounds. That BlueOwl financing of Meta was the biggest private credit round ever.","offset":2700,"duration":25},{"text":"Host: The problem is that private credit in these portfolios has got SaaS businesses and one of the big pitches that these AI compute companies has is that SaaS is dead, which is kind of ironic when ChatGPT on a $20 a month subscription is a SaaS business.","offset":2725,"duration":22},{"text":"Host: So it's kind of funny that they're saying SaaS is dead when they actually are SaaS businesses themselves. But there's been some warnings that have come out around these private credit portfolios that pack in the AI debt as well as the SaaS debt, because AI is actively pitching the future of these SaaS companies will be eroded by the fact that AI is coming.","offset":2747,"duration":26},{"text":"Host: So it's causing some instability in the private debt markets. And of course, what else is causing instability is rising interest rates in an uncertain world and a war in Iran amongst other things.","offset":2773,"duration":18}],"startTime":2626},{"title":"Middle East Conflict's Risk to AI","summary":"The discussion pivots to how the escalating war in the Middle East threatens the AI boom. A prolonged conflict could divert regional investment capital away from tech while massively spiking essential computing energy costs.","entries":[{"text":"Host: Matt, speaking of data centers blowing up, let's talk about the Iran conflict, its connection to AI, and particularly in your latest missive, you express some concerns that there's a risk that potentially this Iran conflict kind of pulls the rug on the whole AI business model. What do you mean, what's going on?","offset":2791,"duration":29},{"text":"Matt Barrie: Well, this is where the Fifth Industrial Revolution meets the Islamic Revolution, right? Um, a lot of the financing for AI has come from the Middle East. And you can imagine now if you're Saudi Arabia or you're the UAE and you have a fiduciary duty to protect your nation and your citizens and your economy, are you going to be putting it into hyper-rounds of Sam Altman's highly inflated valuations, or will you spend it on defense, energy, rebuilding your civilian and industrial infrastructure and potentially going to war to fund an army?","offset":2820,"duration":45},{"text":"Matt Barrie: You know, it's quite problematic when you've got a country that sits right in the middle of, you know, 48 percent of the world's energy infrastructure, that controls a strait where 21 million barrels of oil goes through every day, that can fire $20,000 drones at scale into anything within a 2,000 kilometer range and migrate not just your data centers in the region, but potentially your energy infrastructure into the cloud, literally in a puff of smoke.","offset":2865,"duration":37}],"startTime":2791},{"title":"AI Productivity and the K-Shaped Economy","summary":"Matt argues that AI acts as a powerful tool that exponentially boosts the output of highly skilled workers. He warns this dynamic will fuel explosive entrepreneurship but drastically widen the K-shaped economic divide.","entries":[{"text":"Host: Matt, another theme that you have in your latest missive is that you've got to be pretty smart to really get the most out of AI. What do you mean by that and what are the consequences?","offset":2902,"duration":16},{"text":"Matt Barrie: When I look at, deep down, at kind of what what's happening in the space, I mean, despite you've got a conflation of few things happening right now. You've got the AI companies trying to justify these huge valuation rounds and they're doing that by saying they're going to take away a lot of the world's work. Then at the same time, you've got sort of mass layoffs happening in the market with Block today, with Oracle, I don't know if you saw overnight, but Oracle has announced 18 percent of their workforce has been cut, etc.","offset":2918,"duration":37},{"text":"Matt Barrie: And so you would not be surprised that the general market has kind of conflated these things and thinking that AI is taking people's jobs away. Now, I see AI instead as a tool. It's a very, very powerful tool. It's, you know, but it's a productivity tool, much like the world was, you know, when you went to work and there was no computer on your desk and then you went to work and there was a computer at your desk or you didn't have mobile phones and you have mobile phones or you didn't have the internet and then you have the internet. Yes, there is incredibly disruptive and transformative time and some jobs are lost, but just like pretty much all forms of technology, more jobs are created over time.","offset":2955,"duration":51},{"text":"Matt Barrie: And what I mean essentially by you've got to be smart to use AI is, you know, AI is a power tool and so is a chainsaw, right? You give a chainsaw to a carpenter and they can do, a skilled carpenter, and they can do amazing things. You give a chainsaw to a novice and you can cause all sorts of problems, right? And so what I am observing, both across my platform as well as within my engineering team, is the people who are benefiting most from AI are the ones that are highly skilled and intelligent and know how to use the AI.","offset":3006,"duration":43},{"text":"Matt Barrie: And they're seeing productivity gains which are astronomical. They're seeing, you know, double to triple their productivity and you can measure that in different ways in terms of what they're achieving. And then as you go down the skill level, you do, you know, it is a rising tide lifts all boats. So if you're an average copywriter, you can now be a good copywriter. If you're an average illustrator, you can be a good illustrator using these various tools. If you, I make a joke, if you're an average programmer, well, you certainly are a confident programmer now using these tools. But to really get the best out of them, the people who are highly skilled are the ones that are really, really driving the outcomes. And I see that for example with your work and what you do, you writing these missives and books and so forth. And so I think just like technology has over time created a bifurcation in society where you have the people who are skilled and can create technology and that's where all the wealth flows.","offset":3049,"duration":68},{"text":"Matt Barrie: And that's why you see these companies with huge valuations and, you know, the ultra-high net worths in the technology industry, and then you see a general deterioration at the lower end of society. I think this is going to drive even further to an extent. People who really, really know how to use AI well are going to capture incredible opportunities in all sorts of different market segments where they can control that customer interface.","offset":3117,"duration":30},{"text":"Host: Well, I definitely agree with you. I just finished a writing project, or I should say Claude and I just finished a writing project that was considerably bigger than writing my book Beyond Blockchain. It required a lot more research because Beyond Blockchain was just my own opinions. It was a lot to it. And Beyond Blockchain took me more than three months, this project took me less than two weeks. And it's just amazing how much you can accomplish, although you definitely, as you say, it takes some skill to learn how to manage context window exhaustion and recognize symptoms of it occurring and so forth. So I couldn't agree with you more.","offset":3147,"duration":51},{"text":"Host: It kind of scares me though, Matt, because I think one of the biggest problems that society faces in the mid-2020s is the K-shaped economy. The tendency that the rich get richer and the poor get poorer. It sounds to me like AI really is going to exacerbate that in the sense that the smartest people are going to really benefit in productivity from AI. They're going to be a whole lot smarter and more capable than they used to be. One guy will be able to do the job of 10 or 12 guys. But the 10 or 12 guys that weren't that bright and really didn't figure out how to recognize the symptoms of context window exhaustion in their LLM, I think it really does take their jobs. And it seems to me like this could become the basis for deeper division in society.","offset":3198,"duration":55},{"text":"Host: So, Matt, I want to ask you this, because as the CEO of Freelancer.com, and for any listeners who aren't familiar with Freelancer, it's basically a marketplace where people can hire independent workers, whether they be high-end expert consultants in their field who are the leaders of their field, or if it's just a guy who'll design a logo for you for five bucks. You can hire all those different people on Freelancer. Matt has a very unique perspective on AI because first of all, he's running a company that uses AI, so he leads an engineering team that's building AI-based solutions. But some of the people that his customers are hiring are AI experts that are helping companies to implement AI at a corporate level. And then some of the more mass freelancer, larger group of graphic designers and people that are making logos and so forth, are users of AI that are leveraging their logo design contests and all that. So Matt sees all these different dimensions of different people using AI in different ways, corporations and organizations adopting it, versus an individual guy in Indonesia who's just trying to make a buck as a graphic designer suddenly becoming much more productive and being able to work in different languages that he doesn't even speak. So Matt sees all these different dimensions. Matt, from that vantage point that you have, what would you say are the top three things that you've become aware of that the average person who can't see all that stuff probably wouldn't think of?","offset":3253,"duration":122},{"text":"Matt Barrie: Well, the amazing thing is we're going to see the ability for you to get things done that you could never possibly think of before, right? Rather than just getting a website built, you'd get a whole business built. The ability for, you know, I think we're going to enter a whole new world of explosive entrepreneurship where now you really just need to have an idea to start a company and the ability for you to be able to execute on that using both AI and also access to humans powered by AI will be unprecedented. You'll be able to do it at a cost that's cheaper than ever before. So we're going to enter, I think, into an explosion period of hyper-competition and, you know, thanks to the internet and the ability to distribute products or services over the internet at scale so quickly, if you've got a great idea, that great idea can take off and you can make a billion dollars faster than any time ever in the history before.","offset":3375,"duration":57},{"text":"Matt Barrie: So I think it is an incredible time. I certainly haven't seen as yet the complete replacement of someone in a job. I don't know, you know, if you ask around, does anyone know any graphic designers that completely lost their job? Where the dislocation will occur is where you've got highly paralyzable workflows where you've got thousands of people or hundreds of people doing the same job. So maybe in a call center, where there might be 10,000 people or 1,000 people in a call center doing the same workflow which is customer support, answering a phones, etc. With AI, you'll be able to take it from 1,000 people down to maybe 100 people. You know, if you've got 30 junior lawyers drafting legal agreements in a room, maybe you'll be able to take it down to 13 people in a room.","offset":3432,"duration":53},{"text":"Matt Barrie: But I don't see, because of the way AI works and goes in circles and the failure mode and the fact that as you burn more tokens you've got more chance, even though the unit rate of errors goes down, the fact you're burning all these extra tokens means that ultimately, you know, an error becomes more catastrophic in a way that a human never would make an error. What that means is that I think the ultimate combination is humans and AI. I think it's probably one of the greatest productivity tools known to man and I think it's going to open up a whole amazing golden age of building and creating businesses and hyper-competition.","offset":3485,"duration":46}],"startTime":2902},{"title":"Freelancer.com Innovation and Guest Outro","summary":"Matt shares recent Freelancer.com milestones, including crowdsourcing projects for NASA and the NIH, before wrapping up his segment of the podcast.","entries":[{"text":"Host: Well, Matt, I can't thank you enough for a terrific interview. Before I let you go, you run Freelancer.com, a public company that trades under ticker symbol FLN on the Australian Stock Exchange. Tell people for who are not familiar with it what Freelancer does and how to follow your work. Your latest piece, again, it's linked in the research roundup email, it's published on Medium, it's called Pay to Pray. You write quite a bit of interesting stuff for people who want to follow your work and learn more about Freelancer.com, tell us about it.","offset":3531,"duration":49},{"text":"Matt Barrie: Well, we run the world's largest cloud workforce, so it's about 87 million people in that marketplace. We can do any job you can possibly think of from $10 jobs to $10 million jobs. You know, the mainstream jobs are things like build me a website or build me an app or, basically help me start a company or grow my company. The biggest things we run, we've got a moonshot innovation challenge program at the high end where we help all sorts of US government departments and enterprises around the world solve scientific and technological challenges. So, you know, give us your hardest scientific or technological challenge and we'll solve it or you don't have to pay the prize out. So the biggest thing we've got running right now is the seven and a half million US dollar gene editing challenge for the central nervous system of humans, which we're doing for the National Institute of Health and also working with NASA.","offset":3580,"duration":66},{"text":"Matt Barrie: Actually, ironically, when Artemis goes up in the next 24 hours into space, we worked with NASA to crowdsource the mascot from kids all around the world that's going up with the astronauts to basically inspire kids about space and so forth. If you want to get anything done, you come to our site, we can get it done, and no job is too small or too big or too complex and it ranges through to mechanical engineering or electronic design or whatever you need. And if you want to follow my writing on AI, there's a series of, obviously a podcast that we've done together on MacroVoices on AI, I think this is maybe the fifth or the sixth. And if you go to Medium or Substack you'll be able to find me and see my essays.","offset":3646,"duration":59},{"text":"Host: And if you just put Matt's name in, Matt Barrie, at the search bar at macrovoices.com, you'll see a list of all of the previous AI interviews that we've done right here on MacroVoices. Patrick Ceresna and I will be back as MacroVoices continues right here at macrovoices.com.","offset":3705,"duration":26}],"startTime":3531},{"title":"Iran War Update and Trump Speech Reaction","summary":"Dr. Anas Alhajji joins the show to react to President Trump's latest address. He criticizes the speech for ignoring the realities of the conflict and suggests the world is entering a prolonged global crisis.","entries":[{"text":"Host: Eric, it was great to have Matt back on the show. Now Dr. Anas Alhajji is next on deck for a special second feature interview on the developing Iran conflict and what it means for the oil markets. Then Eric and I will be back for our usual postgame chart deck and trader of the week. Since this extra coverage format seems to be a hit with listeners, we'll do our best to continue it for as long as the situation in the Middle East warrants. Now let's go right to Eric's interview with Anas.","offset":3731,"duration":63},{"text":"Host: Joining me now is Energy Outlook Advisors founder and managing partner Dr. Anas Alhajji. Anas, it's been three weeks since we had you on the program for the first Iran war update. It's been one month since this conflict started. President Trump just moments before we began recording this segment gave an address to the American people. Let's start with the big picture update since this began. How has everything evolved, how have your views evolved since we had you on three weeks ago, and what did President Trump need to achieve in this address and did he achieve that?","offset":3794,"duration":43},{"text":"Dr. Anas Alhajji: Well generally speaking, we got to frame what has been going on and President Trump basically did not frame it. So, either this is about Iran and its nuclear and he tried to make it that way, or Iran part of a bigger picture and the audience basically can choose one of those two. So either it is about Iran and its nuclear program and the outcome of that basically and its impact on the market are completely different from the idea that Iran is part of a big picture. If you look at trade wars, tariffs, sanctions, Venezuela, Panama, Red Sea, China, Greenland, etc., so is Iran and this war is part of it or Iran is completely separate. The president today basically tried to make it about Iran, but all of a sudden he deviated from that and he started talking about we don't need the Middle East, we don't need oil from there, buy oil from me. And this makes the second option basically kind of more credible in this case.","offset":3837,"duration":70},{"text":"Dr. Anas Alhajji: Anyway, we'll go back to that later on. And the other issue is, who closed the Hormuz Strait? You have only two answers. It's either the United States through the insurance companies or Iran. And it's up to you, the audience, basically to believe one of them. There are evidence on both sides. But one of the ironies here is that when we talk about the details of who closed the Hormuz Strait and we say Iran has no power to close it, then people say, well this is a conspiracy theory. It turns out it's a bigger conspiracy theory to say Iran closed the strait. And we heard the president today saying that Iran's navy is gone. Literally said that. And he said they have no radars, he talked about the total devastation of course their leadership is gone, all that stuff. So to believe that we have all these attacks and the aftermath of this devastation that happened to Iran.","offset":3907,"duration":64},{"text":"Dr. Anas Alhajji: We have 28 countries are in charge of protecting the Gulf and the Hormuz Strait through an organization that is based in Bahrain. You have the US navy, you have the Indian navy, you have the British navy, you have the French navy, you have the Saudi navy, you have all the other countries' navies in the area. And then after all this devastation Iran still control the Hormuz Strait? This goes against everything that President Trump said today. So the story does not match in this case. So to sum up, either you look at this war as it's either Iran and nuclear or it is part of a bigger picture of the changes going in the world and then you look at Hormuz Strait and say well it's either the United States and the insurance companies closed it and the main beneficiary of this is the United States and we heard President Trump talking about it today. Or Iran closed it but that does not make any sense and that is a bigger conspiracy theory because how was Iran able to do that after what we heard from President Trump today and the existence of all the other navies.","offset":3971,"duration":64},{"text":"Dr. Anas Alhajji: Unfortunately President Trump today failed to deliver to the nation the message why the war is still going. Many people basically expected that he will announce some sort of an end to the war. We told our clients couple of weeks ago that all the indications point to a long war, and today we got the confirmation. This is not going to end soon. And if it's not going to end soon, there is another question mark about what President Trump said. If everything is destroyed in Iran, then why is this war continuing? What is left to be destroyed? So the impression I got out of this speech is that the people around President Trump are not telling him the truth and he is living in his own ivory tower without looking at what's going on in reality. And we have countries around the world that are literally having blackouts, country do not have propane to cook, countries basically are struggling to get petroleum products, etc. Many countries are worried about food supplies and the president did not even mention it today. This is a global crisis. This is the largest global crisis we have in our lifetime. And the president even did not talk about it and he did not explain to the American people why this war is still going until now. If everything is destroyed why the war is not ending?","offset":4035,"duration":93}],"startTime":3731},{"title":"Leadership and Control Within Iran","summary":"Anas clarifies the internal power dynamics in Iran. He explains that the Iranian Revolutionary Guard has effectively seized full control of the government following recent leadership casualties.","entries":[{"text":"Host: Anas, who is actually in charge in Iran and calling the shots? Because we've seen so much conflicting news in the last few days. President Trump before this speech had said very clearly in his Truth Social posts that what the speech was likely to be about is that Iran had requested a ceasefire and that Trump was likely to agree to that ceasefire. Now preempt, apparently trying to preempt this speech, the guy who is still technically the president of Iran, although the president is not really the head of state who's in charge in Iran, it's really the Supreme Leader who calls the shots. But this guy who is the president issued a letter to the American people in which he says look we don't want any ceasefire, we would not agree to a ceasefire, we want the war to be completely over and we won't accept any end until we have a promise of no further attacks.","offset":4128,"duration":59},{"text":"Host: So it seems like if nothing else that whether that was a true statement that's reflecting the Iranian government or just the guy who's not really in power anymore expressing his view isn't clear, but it seems like it did have the effect of taking the ceasefire discussion out of President Trump's speech. Meanwhile, we're told that who's really in charge of Iran is the Supreme Leader, but that would now be the killed Supreme Leader Ayatollah Khamenei's son. Nobody's seen him since he supposedly took power. Nobody's heard from him. We're told that he's still alive, but we don't know that for sure. Who's calling the shots in Iran and how should we interpret the various messages that we're hearing out of the media?","offset":4187,"duration":53},{"text":"Dr. Anas Alhajji: The reason why we call it the Iranian regime while in other places we call it government, but in Iran we call it a regime, because we have the official government and then we have the support coming from various militias and groups of that government. If you sum them up we call that a regime. So the Iranian revolutionary guard are not part of the government, but once you put them together they become a regime. And what we've seen historically is that there are of course differences between the government itself, the official government where you have prime minister and you have president and you have a foreign minister, they are more diplomatic. While most of the military actions and the attacks and everything else comes from the revolutionary guards who are not part of the government. So historically we've seen that division.","offset":4240,"duration":56},{"text":"Dr. Anas Alhajji: But even within the revolutionary guard they are not really one unit. And what happened is that after the killing of most of the leadership including Ayatollah Khamenei who the religious leader, they decided to at least for a short period of time to work independently. So they told them okay we have no leadership at this stage so every group you are on your own. So we've seen various groups basically acting on their own, attacking on their own, based on their location, based on their beliefs, based on whatever their leaders basically believe about other countries. So we've seen attacks on the Gulf nations, we've seen more attack on some relative to the others, etc. And then the foreign minister came in and said we know nothing about it. Probably he's right because there was a complete disintegration of the system at least for that period of time.","offset":4296,"duration":65},{"text":"Dr. Anas Alhajji: At this stage, it's very clear that the revolutionary guard basically took over the government. So now we can say that unlike in the past where we have a government and then we have the revolutionary guard and the others, now the revolutionary guard basically literally took over the government, so they are one unit right now. And the whatever kind of nice messages we get, we get from people who have no actual power on the ground. And so everything you hear about the Iranian parliament, this is just a joke. This has no impact. So when they say oh we are going to install a tollbooth for example and we got to charge $2 million per ship, etc., this is coming from people who have no power on the ground at all. Unfortunately the western media basically took them seriously and spread the misinformation all over the world. But they have no power.","offset":4361,"duration":62},{"text":"Dr. Anas Alhajji: And speaking about Iran's ability to charge for it, under international law there is no way that they can charge. There is no way they can charge. And whatever they are doing if they are getting any money from anyone, this is an extortion. And the funny part here is if there are governments that are literally cooperating with Iran and paying them so their ships can pass, so we are talking here about Pakistan, we are talking about India, we're talking about Malaysia and probably others, why President Trump is not mad at them? Because they are cooperating with those terrorist and the regime that is killing people who killed 40,000 people of its own people. Why? Because if they are paying them money then they are supporting the regime. Why we did not hear a word about that? Simply because probably it does not exist in the first place.","offset":4423,"duration":59}],"startTime":4128},{"title":"Crossing Red Lines and Escalating Attacks","summary":"The conversation explores the unprecedented nature of the war as both sides abandon established red lines. Anas notes this has led to the nonsensical targeting of regional assets across the Middle East.","entries":[{"text":"Host: Anas, I want to follow up on a really important point that you made in your last interview three weeks ago. What you explained is that most people assume that Israel has nuclear weapons that they could use against Iran but Iran doesn't have a nuclear weapon that they could use in retaliation. You explained that Iran really does have the equivalent of a nuclear weapon in terms of the consequence of using it, which is that Iran could target the desalination facilities of both Israel and other allied Gulf countries which unlike Iran itself which only gets 3 percent of its drinking water from desalination, Israel and other countries in the region are extremely dependent on desalination. So if Iran attacks their desalination they are forcing starvation and just horrible, horrible humanitarian outcomes. They don't have that same vulnerability with their own desalination plants because they only get 3 percent of their own water from desalination. And so you made the argument that Iran could potentially if they really wanted to use the proverbial nuclear option they could escalate to targeting desalination facilities.","offset":4482,"duration":79},{"text":"Host: Now about a week after you made that prediction, Iran made a formal announcement saying if our energy, our civilian energy infrastructure is targeted, we will respond by targeting the enemy's desalination facilities which could bring about that horrific humanitarian outcome that you described. So it seems like this is escalating and escalating quickly. Do I have that right and how do you see this playing out? How serious of a risk is this?","offset":4561,"duration":35},{"text":"Dr. Anas Alhajji: Yes, you are absolutely right. And one of the issues that we have to really realize is when we say this is a historic event, it's not only in term of impact. It is historic in various ways. One of them is there are no red lines. There are no, no one basically has any red lines, the United States, Israel and the Iranians. No red lines. And if you look at the attacks of the other countries in the Gulf etc. there were no red lines. So on every part on the parts of everyone there are no red lines and that make it more, in a sense, another reason to be historic. But one of the other reasons why it is historic because of the nonsense, there are a lot of nonsense going on that we cannot even explain at this stage. We don't know what's going on. For example, Iran is hitting the GCC countries, that's the six countries in the Gulf and some targets in Iraq.","offset":4596,"duration":57},{"text":"Dr. Anas Alhajji: They are hitting them because they said oh you are cooperating with the Americans and the Israelis and therefore we got to hit you. Well if that's the logic then we need to explain this. The Israeli planes that are hitting the targets in Iran and the Israeli rockets basically that are using the fuel, that fuel is some of it is coming from Azerbaijan and Azerbaijan is next to Iran. It's a neighbor. And Chevron is the largest oil company operating there in an oil field. So it is an American company. We saw Iran hitting for example a Ruwais refinery in the UAE. Ruwais refinery is a joint venture between ADNOC and European companies, so there are no American presence there and there are no military bases there. Yet they attacked it.","offset":4653,"duration":55},{"text":"Dr. Anas Alhajji: But why they are not attacking Azerbaijan who is literally giving the oil to Israel to bomb targets in Iran? Same thing for Kazakhstan, the same thing for Turkey because that oil coming from Azerbaijan basically is coming through Turkey. People saying well Turkey is a NATO member. Well I can tell you we know that Trump does not care about NATO and Trump does not care about Turkey if Iran attacks Turkey. But there are, all this confusion and all those issues that do not make sense at all. For example, attacking Salalah oil depots in Oman. Salalah is like, if you are in Iran Salalah is at the other side of the world. And it's an oil depot. Why you are hitting it? There are no Americans there, there are no Israelis there, there's no military base there. And how come literally a drone will go all the way across the Emirates, across Saudi Arabia, across the empty quarter, go all the way to the end of Oman to the other side where Yemen is and hit the target precisely?","offset":4708,"duration":67},{"text":"Dr. Anas Alhajji: So there are too many questions, there are too many things that do not make, do not make sense. For example, hitting this refinery. If Hormuz Strait is closed and the refinery cannot export, why you are wasting ammunition, why you are wasting drones or rockets to hit it while the refinery basically is useless? Why hitting a tanker like the Kuwaiti tanker yesterday that's been in the same place for a month? It's 200 mile away from Hormuz Strait. It cannot go away, it cannot go through the Hormuz Strait, is not even moving. Why hitting it while you know it's not going to go anywhere? And was it a mistake, was it, was the objective something else? What, whatever the case is there are too many questions that have no, no answers. But the bottom line here is the Hormuz Strait is still closed and that is a big failure for the US policy in the region regardless of what Trump claimed earlier in the speech.","offset":4775,"duration":69}],"startTime":4482},{"title":"Geopolitics and the Hormuz Strait Closure","summary":"Anas argues the effective closure of the Strait of Hormuz is driven by a lack of shipping insurance rather than Iranian naval power. He also highlights how China proactively stockpiled energy before the conflict began.","entries":[{"text":"Dr. Anas Alhajji: The fact that the Hormuz Strait is closed, even at this moment, regardless of who closed it, is a failure of US policy. And there are strange things again when we talk about historic event, there are some very strange things. I'm going to tell you some. For example, if we go back to the national security strategy, it was released in November about four months before the war. So I'm going to read this to you from that strategy. So I'm reading right now. \"America will always have core interest in ensuring that Gulf energy supplies do not fall into the hands of an outright enemy. That the strait of Hormuz remain open, that the Red Sea remain navigable, that the region not be an incubator or exporter of terror against American interest or the American homeland, and that Israel remain secure.\"","offset":4844,"duration":62},{"text":"Dr. Anas Alhajji: Let's look at this one because this is kind of really strange statement. President Trump today said we don't need the Hormuz Strait, we don't need the Gulf, we don't need any of that. The strategy said we don't want it to fall into enemies' hand. And he did not say that today. This was one of the biggest failures of the speech. He should have said that. But here is the strangest statement in the strategy. Of course the strategy is really conclusions of a larger document that remains secret. We don't know what's in it. And therefore this, this paragraph, several paragraphs that been omitted. And here is why I'm saying this. The United States historically if you look at various strategies over the last 50 years, always talked about freedom of navigation in those waterways or the choke points. Notice the following.","offset":4906,"duration":54},{"text":"Dr. Anas Alhajji: This is the first time we are, we see a document like this, four months before the closure of Hormuz Strait, saying \"The Strait of Hormuz remain open.\" It never been closed. This statement makes sense if the Hormuz Strait was closed just once in history. So whoever wrote that statement there they must have something before this paragraph, \"open.\" If it wasn't closed in history, why you are using this statement, \"remain open\"? And the other thing is you are talking about Hormuz Strait, why you are tying Israel to it? You look at what we have in actual life, we have Israel basically attacking Iran right now and the Hormuz Strait basically is part of it. But the question here is if Anas Alhajji wrote in June and July, that months away, months before the war, months before the closure of Hormuz Strait.","offset":4960,"duration":61},{"text":"Dr. Anas Alhajji: And people who they can go to our daily energy report and they can go to our newsletter and they can see it. I wrote the 12-day war. In June Israel attacked Iran. China had no problem. Everyone was expecting Israel to attack Iran for the last 20 years and finally they did. But once the Trump administration started attacking Iran, that changed everything. And China was ahead, they knew in advance what's going on. So China basically stopped importing LNG from the United States, stopped importing oil from the United States, they were ahead on Venezuela before the blockade, they were ahead on Venezuela before the death of Maduro, etc. But they been and if you go back to the previous shows we had probably the last three shows or four, we've been talking about how China was building those massive inventories, oil inventories and everything else, in preparation for either war, sanctions or a major interruption.","offset":5021,"duration":62},{"text":"Dr. Anas Alhajji: Now we know. China knew. And China one of the least impacted countries by the closure of the Hormuz Strait right now at least in the short run. Once we look at those events and look at the 12-day war, what I've written was the China got the message out of the stories that been published in major news media outlets and very prestigious outlets. We've seen a number of stories talking about Iran closing the Hormuz Strait. Again we are talking about June. And if you look at those stories you will find something kind of very strange about them because all of them have the same talking point, some of them in the same sequence. That means there is a public relation company basically was directing that. Why they were doing it? So what I wrote at that time was China got the message and the message was the Hormuz Strait will be closed, but it's not Iran who is going to close the Hormuz Strait, it is the United States.","offset":5083,"duration":65},{"text":"Dr. Anas Alhajji: Again this is written in June and July. So we have the November document after that talking about the keeping the strait open and the strait never been closed. And then in June and July we were talking about the United States basically closing the Hormuz Strait. And then we have the insurance fiasco that we've seen until now. I know that many people were pushing against this, this idea, calling it a conspiracy theory etc. and they tried really hard to debunk this idea. Here is the issue that they need really to answer. The issue is this. Even if you send the navy to escort ships out of the Hormuz, ships are not going to move without insurance. If the insurance remains non-existent or extremely expensive, ships are not going to go along with the navy. Because shippers are going to tell Trump or Modi or anyone else who wants to send those navy vessels, telling them look, if I go with you and I get attacked and I lose my ship and I lose everything on it, will you compensate me? And Trump is going to tell them no, Modi is going to tell them no, I'm just providing a service for you you should pay me for it I don't have to compensate.","offset":5148,"duration":73},{"text":"Dr. Anas Alhajji: They are not going to go. They need that insurance to exist and to be cheap enough for them to move. So any way you look at it you say it's conspiracy theory or not it doesn't matter. They need that insurance. So it is an insurance story anyway you look at it. On the other side those who are claiming that Iran closed the Hormuz Strait again Trump told us today that Iran has no navy, has no power, Iran, we heard the whole talk. So either someone is lying to President Trump or Iran has no power.","offset":5221,"duration":39}],"startTime":4844},{"title":"The Risk of Targeting Nuclear Facilities","summary":"Anas warns of severe market panic and potential humanitarian disasters resulting from military strikes against civilian infrastructure, specifically referencing attacks on Iran's Bushehr nuclear power plant.","entries":[{"text":"Host: Anas, there's one more nuclear option that we need to talk about which is so far in human history, no military force has ever intentionally targeted an operating nuclear reactor. In other words the horror scenario, the terrorism scenario of intentionally dropping a bunker buster bomb on top of an operating nuclear reactor to turn it into a dirty bomb of sorts. Now one of the things that has already occurred, and by the way article 56 of the Geneva Convention very clearly prohibits the intentional targeting of any nuclear electrical generation station. You cannot target nuclear electrical generation stations by international law. Except either the United States or Israel, I'm not sure who, has already made at least three attacks, air strikes on the Bushehr nuclear power plant which is Iran's one operating nuclear power plant.","offset":5260,"duration":63},{"text":"Host: Now those were not targeting the reactor. If they had intended to blow up the reactor and breach its containment they would have succeeded, so they weren't trying to do that. They were probably trying to take out the infrastructure and prevent it from generating electricity and supplying it to the grid. But it does seem that they're already violating the Geneva Conventions by targeting a nuclear power plant. Where is this headed? I'm just really concerned about the escalation that could occur from here.","offset":5323,"duration":33},{"text":"Dr. Anas Alhajji: Again this is a historic event for several reasons and one of the reasons why because there are no red lines. And what you described is a red line that's been crossed and is going to be crossed again and again and as a result Iran is going to cross this red line on the other side. Because as you know they already listed the four reactors in the UAE as a target, the stations like you said probably not the reactors because they need kind of special bomb busters or others etc. But they literally put that as a target. So there are no red lines and with a war with no red lines everything is possible.","offset":5356,"duration":42},{"text":"Dr. Anas Alhajji: The issue that I experienced firsthand was when I woke up in the morning here in Dallas and I got flood of messages and emails, people scared to death from oh what if Bushehr basically explodes and would we get nuclear and literally some high net worth individuals basically sent me messages and said should I put my family in a car and leave to Oman or any other country. So people basically being panic and that panic alone could cause massive problem because if you are talking about millions of people taking the highways, the rumor on its own is enough to cause problems let alone if it, if it becomes reality.","offset":5398,"duration":48},{"text":"Host: Well as you said Anas, Iran has announced that if their civilian power generation infrastructure is targeted, they will respond number one by targeting the desalination facilities of neighboring countries which you've described as a nuclear-like option because of its consequences. Number two that they will target the United Arab Emirates operating nuclear power station at Barakah, or the Barakah nuclear power station which is outside of Abu Dhabi. They've already said that will be a target. And then in tonight's, this is Wednesday night that we're recording, in tonight's address from President Trump, he said very soon and possibly simultaneously that the United States might target all of Iran's civilian power generation stations. So it sounds like the event that Iran said they would respond to by targeting both desalination and nuclear power plants is one that President Trump just signaled his intention to escalate to. This is a very scary moment. This feels like, you know, a replay of the Cuban Missile Crisis except this time we actually know about it. Am I exaggerating to say it's that big of a deal?","offset":5446,"duration":72},{"text":"Dr. Anas Alhajji: No, but I think we should be very aware of the fact that today after Trump speech, I am completely convinced that the people around him are not telling him the right picture. The way he described things are far away from reality. Let me give you another example on how far away from trade reality he said drill baby drill. Okay, drilling activities in the United States been down since he came to office. US production been declining in recent months. So he is completely out of reality and I am afraid that the people around him are not telling him the truth. Before this speech oil prices were down. Right now after the speech oil prices are 5 percent higher and they are going up. So the message, the main message of the speech basically is that this is a long war and this is going to continue and the market is pricing it.","offset":5518,"duration":57},{"text":"Dr. Anas Alhajji: And I think by tomorrow morning probably many people who were counting on probably it will be a matter of days is going to change their mind. And that will impact of course the markets whether you talk about oil gas LNG or anything else.","offset":5575,"duration":14}],"startTime":5260},{"title":"Global Energy Markets and Naval Strategy","summary":"The interview concludes with Anas assessing the global energy market implications if the US follows through on President Trump's suggestion to abandon its protection of the Strait of Hormuz.","entries":[{"text":"Host: Anas let's move on now to what everyone's waiting for, which is how do we bring this all together in terms of the outlook for oil prices, LNG, energy markets generally. If the US is basically from what President Trump said very directly in tonight's public address, he said the US will probably walk away from the situation in the Hormuz Strait, leave it to other countries that depend on it. It sounds again as you suggested that maybe President Trump isn't getting the full story from the people around him because of course oil prices are set globally. If the Hormuz Strait is not reopened, energy prices are going to stay high for everyone, including US voters through the midterm elections. So it seems like the story doesn't add up.","offset":5589,"duration":55},{"text":"Host: But it does seem like the president intends to walk away from the situation in the Hormuz Strait once they've finished their military operations against Iran and it's unspecified exactly what that is, but the one thing he did say that was a specific indication of intentions was to target potentially simultaneously all of Iran's civilian energy production capabilities, exactly the thing that Iran has said if it happens is what's going to cause them to target desalination and nuclear power plants. What does this mean for energy prices and what does this mean for global energy markets more broadly?","offset":5644,"duration":36},{"text":"Dr. Anas Alhajji: When President Trump is talking about others basically coming and protecting the Gulf and opening the strait, he means only the people he know and he is talking about European leaders. He's not talking about China. He's not talking about the rest of Asia. So let's be clear about that. China will be happy to bring its navy and and have control of the strait. This will never happen and the United States will...","offset":5680,"duration":25},{"text":"Host: The audio cuts there.","offset":5705,"duration":2395}],"startTime":5589},{"title":"Episode Outro and Disclaimers","summary":"The hosts wrap up the episode with standard podcast legal disclaimers, sponsor acknowledgments, and closing music.","entries":[{"text":"...or sponsors. MacroVoices, its producers, sponsors, and hosts Eric Townsend and Patrick Ceresna shall not be liable for losses resulting from investment decisions based on information or viewpoints presented on MacroVoices.","offset":8100,"duration":15},{"text":"MacroVoices is made possible by sponsorship from bigpicturetrading.com and by funding from Fourth Turning Capital Management LLC. For more information, visit macrovoices.com.","offset":8115,"duration":15},{"text":"[Intro Music]","offset":8130,"duration":15},{"text":"[Intro Music continues]","offset":8145,"duration":15},{"text":"[Intro Music fades out]","offset":8160,"duration":15}],"startTime":8100}],"entries":[{"text":"Host: This is MacroVoices, the free weekly financial podcast targeting professional finance, high net-worth individuals, family offices, and other sophisticated investors. MacroVoices is all about the brightest minds in the world of finance and macroeconomics telling it like it is, bullish or bearish, no holds barred. Now, here are your hosts, Eric Townsend and Patrick Ceresna.","offset":0,"duration":28},{"text":"Eric: MacroVoices episode 526 was produced on April 2nd, 2026. I'm Eric Townsend. We've got another MacroVoices double header lined up for you, and it's going to be a doozy. President Trump gave an address Wednesday night that surprised the markets. So I'm recording earlier than usual this week on Wednesday night about an hour after President Trump's speech, in which he said that if no deal can be reached, the US plans include targeting all of Iran's civilian electric power generation plants, probably simultaneously. That's exactly the red line that Iran has previously said would cause it to retaliate by targeting desalination plants and the UAE's Barakah nuclear power station.","offset":28,"duration":53},{"text":"Eric: Now we know the president's negotiating style is to make dire threats and then walk them back once a deal is reached. So hopefully the threat of targeting civilian power generation infrastructure will come off the table soon. But overall, my take is that this conflict is heating up, not cooling down as the market was interpreting before President Trump's Wednesday evening speech. And the president was clear in that speech in saying that the US would be hitting Iran very hard over the next two weeks if no deal is reached.","offset":81,"duration":37},{"text":"Eric: We're going to kick things off with an exceptional performance by freelancer.com CEO Matt Barrie, who returns as this week's feature interview guest for a full interview on the latest developments on artificial intelligence, what they mean for private credit markets, why the AI business model is setting the stage for an eventual market dislocation on the scale of the 2000 .com bust, why paper-token inference pricing is inevitable and why it will collapse the current AI business model, and much more.","offset":118,"duration":39},{"text":"Eric: Then for this week's Iran conflict update, Energy Outlook Advisors founder and managing partner Dr. Anas Alhajji returns as this week's second feature interview guest, and it's going to run just as long as our first feature interview because there was so much to cover after the president's Wednesday night address. So this is going to be a long episode, but a very important one. So sit back, relax, and expect some really exceptional content from both of our guests.","offset":157,"duration":32},{"text":"Patrick: And I'm Patrick Ceresna with the macro scoreboard week over week as of the close of Wednesday, April 2nd, 2026. The S&P 500 index down 24 basis points, trading at 65.75. Markets put together an oversold bounce, but will rallies fail with this geopolitical backdrop? We'll take a closer look at that chart and the key technical levels to watch in the post-game segment.","offset":189,"duration":27},{"text":"Patrick: The US Dollar Index down nine basis points, trading at 99.55. The May WTI crude oil contract up 1085 basis points to 100 spot 12. The May RBOB gasoline up 439 basis points, trading at 3.09. The June gold contract up 500 basis points to 48.13. Gold's approaching its 50-day moving average, bouncing from its oversold conditions.","offset":216,"duration":48},{"text":"Patrick: The May copper contract up 162 basis points, trading at 5.65. The April uranium contract down 12 basis points, trading at 84.30. And the US 10-year treasury yield up four basis points, trading at 4.37. The key news to watch this week is Friday's jobs numbers, and next week we have the ISM services PMIs, the core PCE and CPI inflation numbers, and the FOMC meeting minutes.","offset":264,"duration":28},{"text":"Patrick: This week's feature interview guest is freelancer.com founder Matt Barrie. Eric and Matt discuss the sustainability of the AI business model, why rising risks and weak unit economics could lead to a shakeout similar to the .com era, and how AI's impact on software and private credit could create broader market stress. And stay tuned for a special follow-up with Dr. Anas Alhajji where we break down the escalating Iran conflict and its implications on global energy markets. Eric's interview with Matt Barrie is coming up as MacroVoices continues right here at MacroVoices.com.","offset":292,"duration":48},{"text":"Eric: And now with this week's special guest, here is your host, Eric Townsend. Joining me now is freelancer.com founder Matt Barrie. As usual, Matt has written an excellent paper about AI. This one is called \"Pay to PrAI,\" that's pay to P-R-A-I. You can find that linked in your research roundup email. If you don't have a research roundup email, it means you're not yet registered at MacroVoices.com. Just go to our homepage, MacroVoices.com, look for the red button above Matt's picture on the homepage that says \"Looking for the downloads.\"","offset":340,"duration":43},{"text":"Eric: Matt, before we even get into all of what's going on in AI, we've got some news just as we're recording this on Tuesday evening US time. There's some recent news just in the last day or so, which is, you know, a lot of companies before they IPO, they'll do a little bridge round, maybe 10-20 million bucks just to cover some expenses until they get to their actual IPO. OpenAI is planning an IPO. They did a little bridge round, 122 billion with a B dollars. All-time record for a private fundraising round. And what is that, something like 100 times bigger than any private fundraising round has ever occurred like before 2025. Why did they need 122 billion just to bridge them from now to their IPO, which is expected in less than a year?","offset":383,"duration":45},{"text":"Matt: Thanks for having me. It truly is stupendous. I mean, before 2025, the largest sort of private venture rounds were in the single-digit billions. The headline number is 122 billion raised on a 730 billion pre-money valuation. When you get down to the actual segmentation of kind of what's going on, it seems that it's only about 25 billion dollars worth of cash. It seems to be more of a vendor financing than an actual straight cash injection.","offset":428,"duration":33},{"text":"Matt: You've got Amazon, Nvidia, and SoftBank primarily putting the money, or the in-kind, in. Amazon's putting in 50 billion, but that's contingent on OpenAI spending 100 billion, I think over the next eight years on their compute, which I'm sure won't make Microsoft happy, and it seems like a good deal, at least on paper, for Amazon. So there's 15 billion going in up front and 35 billion is furthermore contingent on the company either going public by 2028 or achieving artificial general intelligence. I'm not sure how they're going to really define that. I think the underlying definition is a panel of experts will make a decision, yes or no, which is a bit strange for a financial decision.","offset":461,"duration":33},{"text":"Matt: There's 15 billion down up front, 35 billion contingent on going public or AGI for a 100 billion commitment the other way around, so it's a bit like a procurement round. SoftBank loves doing this sort of, I like to call them Russian stands, where they kind of lead rounds and mark up valuations to the moon. We saw it with WeWork, we're seeing it again with Enron, I mean OpenAI. They've already put in about 40 billion, they're putting another 30 billion in. But to raise that money, they don't actually have the cash on the balance sheet, so they've taken a 12-month bridge loan for 40 billion and they're tranching in 10 billion at a time over the course of the year for a total of 30 billion.","offset":494,"duration":37},{"text":"Matt: And obviously that's kind of just getting them through to the IPO so they've got a liquidity event. And then Nvidia is putting in in-kind, as they tend to be doing in all these sort of circular economy sort of deals in the AI space, where GPUs and infrastructure will be provided to the tune of 30 billion into the round. So it's about a 25 billion dollar round of cash sort of up front, 10 from SoftBank, 15 from Amazon. We'll see if the rest of the money comes in, but the rest is in-kind.","offset":531,"duration":27},{"text":"Matt: So it seems from looking at this, and if you look at the compute numbers, I mean they truly are astronomical that are being contributed in the form of, you know, either Nvidia credits, I think there's like three gigawatts of inference and two gigawatts of training capacity as part of this investment. You know, that's sort of the power that gets drawn by a small country. So it seems to be what they're trying to do is scale up the spend on compute so much that they can find a way to bring the unit economics down because that's really the key problem in the space, that nobody in the AI compute space is making any money other than really Nvidia who does about 160 billion of revenue and 100 billion of earnings, and then TSMC who provide the chips to Nvidia, but the rest of the space is actually negative on using the product in terms of the unit economics. So the more you use the product, the more you lose the money. So I think they're trying to make it up on volume.","offset":558,"duration":49},{"text":"Eric: Matt, I can't imagine any responsible business executive signing off on 122 billion dollar deal unless the underlying business model was rock-solid and it didn't have any major risks in it. Just happens that you wrote this missive over the last couple of weeks called \"Pay to PrAI.\" I don't think that's actually the conclusion that you reached. Tell us about the economics of the, I'll call it the consumer AI business model, offering AI through chat to people like me who sign up for a Max subscription on Claude or a Pro subscription on OpenAI. How much money are they making or losing on that?","offset":607,"duration":37},{"text":"Matt: The AI industry is consuming an absolute bonfire of money. It's about 600 billion dollars a year that's being spent by the hyperscalers on CapEx. It's reaching a kind of a point now which is kind of incredible where, you know, the CapEx is higher than their internal free cash flow. You know, the fundamental business model that's being pushed in the consumer market and the software development market up until now has really been a venture capital style subsidized model where you pay, you either use the free product and then hopefully upgrade to the 20 dollar a month product, or you use the 20 dollar a month product if you're a consumer or a 200 dollar a month product if you're a power user or if you're starting to do programming.","offset":644,"duration":38},{"text":"Matt: But the problem with these models are that they take an incredible amount of money to train, and I think we've talked about that in previous episodes where you've got training runs North of 100 million dollars a run, you know approaching half a billion dollars a training run. That requires a huge amount of data center build-out, a whole lot, an incredible stupendous amount of data needs to go into these models. And so, you know, because the cheap data that's scraped off the internet for free is basically sort of, you know, drilled out to an extent, they have to do licensing deals to get access to that data. Sometimes they do dodgy things like Anthropic, you know, scraped a whole bunch of books and scanned them in and then they got caught and they had to pay the biggest, I think, fine in copyright history, I think it was about 1.5 billion dollars for the illegal scanning of all that data, etc.","offset":682,"duration":42},{"text":"Matt: So it's incredibly expensive to train. But the fundamental problem is that when you actually use the inference, you basically put queries into GPT or queries into Claude and you run them, those queries are loss-making. You can't make it up on volume under the current models. While we do see, you know, the underlying hardware is on sort of a Moore's Law sort of trend, and Moore's Law for those of you that don't know what it is is, you know, every 18 months to two years or so, effectively the technology that goes into chips allows semiconductors to be produced with finer and finer feature sizes, which effectively allows the compute capability in terms of processing power to effectively double or the cost to half every 18 months.","offset":724,"duration":44},{"text":"Matt: While you are riding on that sort of silicon Moore's Law trend, because the models are in such a brutally competitive environment where there's literally zero lock-in, there's almost zero switching costs, so I can, if one day ChatGPT 5 comes out I'll switch to that, but then Claude 4.6 opus comes out and that's better, I'll switch to that, and so there's nothing stopping me overnight really just changing the models. Because of the competitive environment and so forth, what the situation basically is that each generation of model is being rushed out to kind of get the top of the scoreboard so that all the customers flock to that.","offset":768,"duration":35},{"text":"Matt: And as a result of that, the actual amount of inference or tokens burned, if you will, per useful query with each generation of model, while the underlying compute is getting cheaper and cheaper in terms of what you can get done on the chips, the amount of inference you have to burn for a useful query is actually going up quite dramatically. And so you're actually not seeing a reduction in the cost of inference, you're seeing an increase. And that's before you consider the issue with energy and constrained build capacity for building data centers and power gear and all the other things.","offset":803,"duration":36},{"text":"Matt: And so, you know, a few people have done various models of how much it costs for GPT to deliver their 20 dollar plans, and even if you ask Claude itself, you type a query into Claude saying how much is the underlying compute costs for on a 20 dollar plan with Claude, it will tell you, you know, 15 to 20 dollars, maybe 18 dollars, maybe a bit more. So effectively there's no money being made on these 20 dollar plans and when you're a power user you can burn up to several hundred dollars on a 20 dollar plan. And it gets even worse on these 200 dollar plans which are used by programmers now, because the nature of programming is you're streaming tokens almost forever, and you know, the average user on a 200 dollar plan can burn, you know, many thousands of dollars of underlying compute, and in fact there's a, there's a leaderboard called VibeRank where people compete to see how much underlying compute they can burn on a 200 dollar plan. The leading guy on the leaderboard has burnt 51,000 US dollars in a single month on a 200 dollar plan.","offset":839,"duration":60},{"text":"Matt: So the issue is that as these models get more and more competitive and new versions come out, you know, you're getting an exponential increase in the amount of inference you've got to burn, and that is meaning that these models are not getting cheaper and you're not making up on unit economics. And so it kind of feels like this funding round, especially with the amount of vendor financing that's packed into it, it feels like this is an attempt to really try and scale up the underlying infrastructure and GPU capability so that potentially you know, some sort of threshold can be crossed in the underlying economics so inference could be profitable. But there's one big problem with all that, which is is the demand there? And and that's where we are.","offset":899,"duration":44},{"text":"Eric: Matt, you said it's 122 billion dollar capital raise on a pre-money of 730, so I get 852 billion is the current enterprise value. Assuming an up round, which is what everybody assumes, we're going to go just let's call it an even trillion for the IPO because hey what's, you know, 100 billion here, 100 billion there, you know, eventually you're talking about real money. But holy cow, let's say that OpenAI goes ahead and IPOs sometime this year for 1 trillion dollars. Are you going to be long, short, or flat, and why?","offset":943,"duration":37},{"text":"Matt: It really does feel in the space that we're really in the moment where a supernova is starting to explode and we're probably going to end up with a giant black hole like we did in the .com boom the first time around. I mean a trillion dollar valuation at IPO is absolutely gigantic. You know, they've raised 122 billion here in this round. They were mooting that the IPO earlier in the year was going to be a 60 billion dollar raise on about a trillion, but you could probably imagine that that number has possibly scaled up a little bit or they want to keep it tight obviously because once it gets to public markets you don't want to have too much stock unload onto the market.","offset":980,"duration":36},{"text":"Matt: But a bigger problem for them is the fact that you've got Elon Musk in the wings who's not just suing OpenAI because it turns out you can't IPO a charity and they've converted OpenAI to a for-profit model and there's a lot of complications around that, but he's also IPOing his SpaceX for I think it's 1.75 trillion valuation. So there's a big possibility that a lot of the heat's going to be taken out of the market when he does his 75 billion on a 1.75 trillion dollar raise, and it looks like he's going to beat OpenAI to to going public because if you kind of look at the news and you know what's all happening with the ETFs and the allocations and so forth, he's a lot further along.","offset":1016,"duration":40},{"text":"Matt: I don't know, in the olden days, you know, if you were to invest in Amazon when it went public or Microsoft when it went public, you know, you had the ability to make a lot of money. I don't know how much is left on the table for the general public when the company's being valued at a trillion dollars when it goes public, you know, what are you going to do? Get to two trillion, 10 trillion? I think Nvidia is what, four or four and a half trillion valuation, and they're the only ones making money in the space.","offset":1056,"duration":26},{"text":"Eric: Matt, the parallels between this and the late 1990s .com boom before the 2000 .com bust are just striking to me. And it occurs to me before I go on that we probably have listeners that weren't even born when that happened. So for anyone who's not familiar with what happened there, Wall Street became absolutely obsessed with the idea that the internet is going to be a really big deal. And the thing that's really important to understand about this is they got that call exactly right. The internet, the public internet was going to change the world we lived in in ways beyond what anyone could even conceive and they got the call right that it was a really big deal.","offset":1082,"duration":44},{"text":"Eric: And I think they're getting the call right again that AI is a really big deal, maybe as big or bigger than the public internet. But the thing is, even though they got the call right, they started throwing money at dumb ideas without thinking and it just turned into a complete frenzy. It seems like that's happening again here, but in the late 90s, it was every tiny little company that had .com in its name and it didn't really matter whether they had a business model. It's different here. It's the big players, which frankly have a business model. But as you very eloquently explain in this excellent piece I recommend everyone read called \"Pay to PrAI,\" that business model isn't viable for the reasons that you just described, or it's not profitable or not likely to be sustainable.","offset":1126,"duration":50},{"text":"Eric: How should we think about this? Is it inevitable that a .com bust like we had in 2000 is coming for AI, and does it have the same dimensions as the one in 2000? Is it bigger, smaller, worse, better? What do you think?","offset":1176,"duration":16},{"text":"Matt: So if you think about the .com boom, and I was there actually in Silicon Valley in '97, '98, '99, 2000, I saw it all. There was a hypothesis the internet was going to be a big thing and it turned out to be a enormous thing in terms of the benefits to humanity and society and will continue to grow in terms of its applications. And AI is the same thing. AI is going to be absolutely transformative for humanity in terms of what it can do. At that time, you know, a company whose valuation went through the roof was Cisco, and their tagline was \"We network networks.\" We talked about it I think last MacroVoices. Every time you plugged in a bit of the internet, you needed to have a router to connect up the network and Cisco equipment was going to be everywhere and why wouldn't it be the most valuable company in the world, right?","offset":1192,"duration":42},{"text":"Matt: And the same as with OpenAI in a to an extent, you think okay, well they've got the best AI models, AI is going to be everywhere, why wouldn't OpenAI be the most valuable company in the world? But then at the same time, you also had AT&T, and up until about 1996 AT&T had a 60% market share. It basically built the network using Cisco equipment to connect up the world. You had three players in the market and I think it was mentioned at the time by an analyst that they had margins that would make drug dealers blush.","offset":1234,"duration":25},{"text":"Matt: But the problem was that in 1996 the Telecommunications Act, Regulation Act came in and deregulated the market. And you started having fourth entrants and fifth entrants and so forth and and people buying and selling capacity to each other etc. and when you started having the fourth entrant come in and competing on cost, the unit economics fell apart. And if you think about the AI compute space, well Amazon had it pretty good up until the AI boom, I think its its CapEx as a percentage of earnings was down to about 6% at one point.","offset":1259,"duration":30},{"text":"Matt: Now these hyperscalers are spending, or the cloud computing companies are spending, 60% of earnings on CapEx, over 100% of earnings on CapEx. They're spending 600 billion dollars a year at the moment in terms of a run rate in CapEx and it's hypothesized that by 2030 it's going to be 5.2 trillion dollars of CapEx, right? Now when you just had Amazon, you know, really dominating in terms of of cloud, you know, it was sitting pretty and I think it had about 60% market share.","offset":1289,"duration":26},{"text":"Matt: Now you've got Microsoft who's taken over and they're market leader and so forth. Now you've got entrants such as Oracle that's competing on price and you've got CoreWeave and the Neo clouds and the unit economics are starting to look a little bit shaky. And then you had a big bust in the .com bust where, you know, all these companies had money being thrown at them and you're seeing that right now. Any company that kind of has AI in its business plan is getting these stupid, you know, seed rounds in the hundreds of millions of dollars. I remember when seed rounds were hundreds of thousands of dollars, now you have hundreds of millions of dollars.","offset":1315,"duration":36},{"text":"Matt: And you know through the early 2000s you had a real trough in the technology space where where everything blew up and was somewhat uninvestable for a period of time. But through that period of time, you had companies like Google and Amazon continue to grow and double down and Microsoft and so forth and ultimately today became very very big companies. So I think we're seeing this on steroids right now with AI CapEx and compute and the OpenAI funding rounds and so forth. And you know it's unquestionable AI is going to be completely enduring and just absolutely game-changing for society. But I think the kind of circle jerk of money that's sloshing around between a very very small number of companies at stupidly high valuations and at a scale that is just stupendous is going to potentially end up in a big bust.","offset":1351,"duration":51},{"text":"Eric: Matt, I want to focus on that because a lot of people are going to mishear what you just said and they're going to say Matt Barrie said imminently tomorrow there's about to be a great big bust in all of the AI stocks. That's not what you said and it really rings home for me because the reason that my partners and I sold our software company in the summer of 1998 is because we knew it was a bubble. And my plan at the time was to take all of the proceeds and short the NASDAQ because I knew it was a bubble.","offset":1402,"duration":33},{"text":"Eric: Fortunately I got talked out of that, but if I hadn't, I would have lost everything because even though we were right, basically the NASDAQ doubled between '98 and 2000 before it crashed exactly like I predicted it was going to crash. What do we do here? I mean, do you go long, do you go short? And I'll say I'll ask the question this way: OpenAI and Anthropic are both expected to IPO probably in 2026. When they do, you know, is it time to go long or is it time to go short? Because you know I could make either argument. It seems to me eventually you want to be short, but how do you time that?","offset":1435,"duration":37},{"text":"Matt: That's the trillion dollar question, right? You've got SpaceX going public which obviously has Grok and xAI within it. You've got Anthropic going public and you've got OpenAI going public and that's why I think it feels a bit like a supernova, so we'll end up with a big bang in one way or or another. I mean, you know, I think the issue is going to be, I mean that some of these valuations like the OpenAI valuation is kind of predicated on the fact that it's going to capture an enormous amount of value out of the world in order to justify these valuations, right? You know, these funding rounds are so large and the valuations are so high that you need to pitch a revenue line that matches them.","offset":1472,"duration":36},{"text":"Matt: And you know at the moment OpenAI is doing about 2 billion a month in revenue, I think it's at about a run rate of about 25 billion, I wish they'd stop using the word run rate because I think you've actually got to recur once before you can call something annual recurring revenue. So it's doing about 2 billion a month of revenue, losing you know 14 billion this year, I think the burn rate that came out today was is expected to lose 70 million dollars a day this year scaling to 156 million a day lost next year. But in order to to, you know, cross the valley of death and get to the promised land, they've got to show a business model that makes sense.","offset":1508,"duration":34},{"text":"Matt: And I think what their business model is going to is twofold. One is that they're going to take a substantial amount of white-collar jobs away from humans, you know some substantial percentage of jobs in the world are going to go to AI and that's kind of the top-line revenue number. And then the in terms of generating earnings out the other side, the justification is that the infrastructure that's being contributed as part of this 122 billion dollar round is at such a scale that only OpenAI will have the unit economics that will make the inference profitable. So in a combination of taking everyone's job and having the only infrastructure that can run this stuff profitably, I think that's the argument for justifying these sort of trillion dollar valuations.","offset":1542,"duration":42},{"text":"Matt: Now I don't think OpenAI is going to capture, for example hypothetically, you know the value in the AI-powered drug discovery market anymore than AT&T captured the value of the iPhone, right, you know? The underlying technology phenomenal, but as Ilya Sutskever said himself, who was one of the founders of OpenAI, that you can just read 40 academic papers and 95% of what's out there in AI is is published in in the public domain. And that's why every couple of weeks there's a kind of a new foundational model that's top of the leaderboard and every once in a while there's a team of people that you've never heard of, that 160 engineers in a room in Hangzhou like like DeepSeek that comes out of nowhere and suddenly leads in the leaderboard because it turns out that there's no sustainable competitive advantage in the underlying foundational models and what you do need though is access to data.","offset":1584,"duration":44},{"text":"Matt: Those data sets are now kind of out there in public or in the case of the Chinese they probably don't care too much about copyright so they'll always have a sustainable competitive advantage. And I think there's incredible and tremendous opportunity in AI, but it's probably not going to be in these models that that haven't really figured out what the business model is.","offset":1628,"duration":23},{"text":"Eric: Matt, I think there is a very distinct difference between this scenario and the late 90s to 2000 .com story, and that is this. Let's say that we don't know whether it's '97, '98, or '99 right now, but we know that March of 2000 is coming at some point. There's going to be a washout in this commercial AI space where they just realize that the business model wasn't sustainable. Well what happened in the .com bust is we went really a good solid couple of years of kind of a technology recession where there wasn't a whole lot of progress on the public internet because we had to basically shake that malinvestment out of the system, get back to efficient capital allocation, and then you know after 2002, 2003 things started to really take off again.","offset":1651,"duration":51},{"text":"Eric: What if the US government steps in and says, now wait a minute, the military applications of AI create an existential threat to the country if we don't stay ahead of this? We can't tolerate a 2000 to 2003 pause. You must continue, you must do it under US government funding, but we're not going to fund the giving away stuff for free to Max subscribers on Claude. We're going to take it over and you know it's just for the military now. We're not going to have consumer AI anymore. Is that a realistic scenario, or do we need consumer AI in order to train the models in order for the military to get the benefit, you know, what would happen in that scenario where the military says no, we we can't allow a stock market crash to slow down progress on the military applications of AI?","offset":1702,"duration":49},{"text":"Matt: So I don't think it's realistic that we're not going to have consumer AI. I mean the Chinese are open sourcing their models and in fact they've got their strategy to open source both the software and the hardware. You know, and there's been previous leaks of Meta's Llama and and so forth. I think the interesting thing here is that it won't be the government stepping in. I think there's one big company that's kind of been sitting in the wings that has a God-awful AI product and it's in everyone's pockets and that's Apple and the iPhone.","offset":1751,"duration":25},{"text":"Matt: I think in our last chat we talked about how goddamn awful Siri is for supposedly being your original chatbot AI assistant and we hypothesized that it will continue to be awful in the next year and here we are and it it continues to be pretty bad. But they've kind of avoided this whole getting sucked into this whole CapEx bonfire and they've really got two levers they can pull. One is is they're sitting on this cash and they've kind of they can sit back and watch and and as with the .com bust, there was a lot of infrastructure that went through to the second and third owners where you know maybe the first owner that builds out the optical fiber network goes bust and the second owner comes in and tries to make it break even after after purchasing it at a at a cheaper price and washing out the underlying sunk costs, and then the third owner comes in and makes some money out of it.","offset":1776,"duration":46},{"text":"Matt: And I think you might have that situation in the AI compute space where if there is a problem with some of these large foundational model companies, there might be a second or third owner and Apple would be ideal for that. At the same time, what they've been doing is they've been putting in some AI silicon into their products and a lot of this compute that's currently being done in a data centers, you know, by OpenAI, by Anthropic, etc. is going to go to the edge and it's going to go to the edge for a variety of different reasons. One is that you don't want Sam Altman training on your data, and I think we've talked about before that I think an emperor has no clothes moment is heading into into SaaS where you know I think large enterprises are starting starting to think to themselves, you know what, I don't want my data in Google Drive and Gmail, you know it might get trained on.","offset":1822,"duration":44},{"text":"Matt: And you're certainly seeing all the major SaaS companies quietly flicking on a switch in the settings without telling you where they're saying by default now we can train on your data potentially, and then you flick it off but by then it's too late and all your data has been sucked down. So a lot of that compute is going to go the edge, it's going to go to on-prem, you know in the enterprise on compute devices so it doesn't go into the cloud. It's going to be on your phone, it's going to be on your MacBook, it's going to be on your Mac Studio or whatever it may be for privacy, for confidentiality, for latency reasons and the fact that there's some models now that can actually fit on that silicon.","offset":1866,"duration":34},{"text":"Matt: And you know while it might be you know one or two generations away, you know at some point I think real soon a lot of that load and a lot of that compute that's going to AWS data centers and Azure data centers to run Anthropic and to run OpenAI is going to go to the edge. It's going to be on Apple's devices, it's going to be on Apple's laptops, it's going to be on-prem, at the same time Apple's cashed up and might end up being the second or third owner of some of these companies.","offset":1900,"duration":26},{"text":"Eric: Matt, the title of your missive that you just written, \"Pay to PrAI\" or \"Pay to PrAI\" P-R-A-I is actually a reference to the inevitability in your prediction of something called paper-token monetization. What does that mean, how does that relevant, explain what that's about?","offset":1926,"duration":19},{"text":"Matt: Well, the fundamental business model of Silicon Valley venture capitalists is to try and win markets by financing companies with astronomical amounts of money such that that money gets spent on marketing and subsidization of the product to such a scale that nobody can compete with these Silicon Valley invested companies or unicorns. And so total nuclear war is launched on a market, so you can think maybe Uber and free rides in China or whatever it may be or you know DoorDash delivering noodles to people in Indonesia or you know Grab or whatever whatever it may be.","offset":1945,"duration":35},{"text":"Matt: So what we have here in the AI space, the subsidization model is that you have a free product being GPT or what have you, and then you've got a 20 dollar a month product which I've said earlier probably costs them 20 dollars to serve, and a 200 dollar a month product which probably costs them 2000 dollars to serve. You just can't continue forever. At some point, so the so the question is going to be can we get the unit economics down to do inference to such a point that these models are profitable before these companies run out of funding?","offset":1980,"duration":30},{"text":"Matt: Now the problem is that as we've moved into for example software development which consumes a never-ending amount of tokens to write code because every company in the world is powered by software now, you know there's this huge token burn that's happening in these 200 dollar plans. And if you were to try and make some sort of reasonable software-like margin, like 80% etc., you would have to price these plans not at 200 dollars a month but maybe a thousand or two thousand dollars a month or higher.","offset":2010,"duration":30},{"text":"Matt: At some point the money is going to run out and I think they had a near-death experience in the last couple of weeks. Anyone who and I know you're constantly using GPT and Claude etc. and you probably noticed the same thing, you know in the last couple of weeks there seems to have been a bit of a panic from these companies where you know you log into your 200 dollar plan, you type a couple of queries and then you run out of credit. You've always known that these companies aren't making money on the inference because instead of saying put your credit card in and top up your credit, it puts you in the naughty corner for seven hours or longer.","offset":2040,"duration":31},{"text":"Matt: The inevitable destination for these subscription models which are basically massively subsidized Silicon Valley financed and increasingly debt-financed business models is that they're going to have to move to a per-token pricing. Now and that is going to cost a lot of money. I think there was some comments in the last couple of weeks on Reddit where someone who's on a 200 dollar plan ran out of credit pretty quickly, he had to get it done he had to get something done you know that night, and so he moved to the API pricing which is in the programming interface and that was costing 200 dollars an hour instead of 200 dollars a month.","offset":2071,"duration":34},{"text":"Matt: And he thought gee, I've just you know used a couple hours used 5, 600 dollars. The problem with that sort of pricing has several dimensions. The first is that already at 20 dollars a month for a plan which is 240 US dollars a year, that already prices the product out from over half the world's population. I mean the median global income in the world is about 2.5 thousand dollars a year so at 240 dollars a year you're already 10% of the pre-tax income for half the people on the planet, right? So you're already quite expensive.","offset":2105,"duration":33},{"text":"Matt: The second point problem here is that you know in these programming models which which realistically to make any sort of margin need to be priced in the thousands of dollars a month or maybe 10,000 dollars a month, you have a bit of a problem in the fact that you know these models do hallucinate. And when they do hallucinate and throw errors, they're very different from human. When you use hire a freelancer for example on my website freelancer.com, you know you put in some money, okay I'm going to pay you 200 dollars to build a website for me and you don't release that milestone until the job is done.","offset":2138,"duration":33},{"text":"Matt: And if the human can't figure out a problem, you know they'll they'll ask their friends, they'll try and find a more senior engineer for advice, they'll get on a forums and ask other people how to solve problems, they'll browse the internet, they'll they'll kind of hill-climb their way out of solving problems. And look, they may they may ultimately not get there and you might get frustrated with them and want to find another developer to do the job, but ultimately their failure modes are very different from AI. With with AI, sometimes when it hallucinates, it can do wildly crazy things.","offset":2171,"duration":28},{"text":"Matt: About three weeks ago, I had a problem with my with my VPN on my computer, I was using Claude to kind of help me through figuring out how to get it fixed. I'm a software developer by background, I've got electrical engineering degree from Stanford, you know I I do know how to program quite well, but I kind of veered into PowerShell commands in Windows 11 which I don't know very well and I was blindly pasting in Claude and it made me delete my entire networking stack.","offset":2199,"duration":26},{"text":"Matt: And so you have these very crazy failure modes with AI where you either go in loops or it goes away and thinking and and you know burns you know 10x the inference you know try you know these reasoning chains trying to figure out what's going on, or it just has these crazy suggestions and and it will look at you in the eye with the the eyes of a sociopath on a first date in some regards trying to you know gaslight you into thinking that its answer is is true when it's clearly not the case, it's just hallucinated something or rather.","offset":2225,"duration":32},{"text":"Matt: And the issue here is that in a paper-token pricing model, it really turns software development into a slot machine in that if the if ultimately you know you're writing your app and you you kind of need to get Claude to do something or rather, you're really pulling the slot machine handle, you don't know how much it's going to cost you by the time the tokens are all burned and you don't know if you're actually going to get a solution at the end of the day. And so, you know I guess only in Silicon Valley could they turn software development into degenerate gambling because that's that's kind of where you end up.","offset":2257,"duration":33},{"text":"Matt: And the frustrating thing I think will happen is when you're on the 200 dollar plan you've got a certain amount of capacity and maybe it tells you to time out or what have you, you kind of know, you're capped at 200 dollars, right? It's very frustrating what it runs out of credit you gotta find alternative ways potentially of doing things, but you kind of know what you're what you're up for. When you're pulling the handle under the paper-token model, it could be 50 dollars per spin, you may go in a circle, you may go in a circle 10 times, you know there's all these examples of people get over X of of radical crazy things that that that Claude does to your codebase.","offset":2290,"duration":36},{"text":"Matt: And I think people are going to get very very frustrated if they have to kind of put a coin in the machine, pull the handle every time and they get a non-deterministic outcome of whether they're moving forward in a hill-climbing sense to to their final solution and their final app or their final bit of software being developed or whether they're going in circles you know round and round and round round again. And I think people are going to get frustrated, they're going to go try and find open source models, they're going to try and find alternative ways to get things done and I and I think at that point, that's where the bubble starts to pop, it's really the problem.","offset":2326,"duration":40},{"text":"Matt: And while the hallucinations are reducing as part of you know each new generation in terms of the engineering of the infrastructure around the foundational models to reduce those hallucinations, what is actually happening is the token spend is going up exponentially because you're doing more and more complex things, so you have a much bigger surface area in which you could generate an error. So what I mean by that is, you know, while the probability of a failure pulling the pulling the handle on the slot machine is getting reduced with the you know the engineering from each subsequent generation of model, the number of handle pulls you need to make is is going up exponentially and so you've got a multiplicative effect in terms of potentially the impact of errors.","offset":2366,"duration":30},{"text":"Matt: I just think it's going to, the question that basically needs to be solved now with this 122 billion dollar fundraise is can you get the cost of compute way down to get this whole model profitable and will the market tolerate software development as a slot machine.","offset":2396,"duration":6},{"text":"Eric: A lot has been said already by lots of people about the incredible rate of progress and how quickly AI itself is getting smarter. What I don't think has been discussed enough and I'd like to get your comment on is the rate to which professionals are becoming dependent on it. I think more than you know this it's more addictive than cocaine. I remember our first interview on AI when ChatGPT had just come out and I remember thinking to myself, boy Matt's really into this stuff, to me it's a novelty but I don't think I'd ever pay 20 bucks a month for it. I mean it's just it passed a Turing test, big deal but I don't think I'd ever buy a subscription.","offset":2402,"duration":46},{"text":"Eric: I'll tell you Matt, in the last several weeks, there's been a lot of stress in my life because of this war and family that are affected by the war and waking up double digit percentages down on a percentage basis on my net worth because of something that happened in the market overnight. Okay look, I'm a big boy I've been through that stuff before, I've been through the 2008 crisis, it's not that big of a deal. But Claude 4.6 was going offline and the server wasn't available and I was freaking the F out. I couldn't handle it. I was losing, and don't you dare insult me by suggesting that I go back to ChatGPT 5.4. I don't drink Pabst Blue Ribbon and I don't do ChatGPT, okay? I mean I've gotten to the point where I can't live without Claude. This is this seems like a risk.","offset":2448,"duration":55},{"text":"Matt: When Claude went down, it's quite interesting actually because the first time it had a bit of an outage was when the data centers got blown up in in Dubai and you kind of have to wonder to yourself what potentially was being run in those data centers in the Middle East that caused the outage of Claude. And I do think those companies did have a bit of a near-death experience in the last couple of weeks. Anyone who and I know you're constantly using GPT and Claude etc. and you probably noticed the same thing, you know in the last couple of weeks there seems to have been a bit of a panic from these companies where you know Sam Altman killed Sora which was this hyped-up video modality model where you could generate you know clips or you know the whole point was you were supposed to be able to type in a prompt and get a movie out the other side.","offset":2503,"duration":31},{"text":"Matt: In fact Disney paid a billion dollars to OpenAI for use of the technology and only found out half an hour before a meeting that the whole thing was going to be canceled. At the same time Sam Altman killed instant check-out, he's working on some sort of erotic model as well which is a bit strange but I think it's probably one of the big markets is pornography and he thought maybe we could make some money there. He killed that as well and at the same time Claude had these big changes in terms of how they did the plans and they were giving out you know extra tokens in off-peak but in peak they'll kind of cutting you back etc. and then ultimately it seems that they've they've cut back a lot of lot of the token budget you get in these models and on the path to this sort of paper-PrAI paper-token sort of business model.","offset":2534,"duration":40},{"text":"Matt: So I do think they all had a bit of a near-death experience and that near-death experience obviously is what I was feeling or what you've been feeling when you kind of see these models you start using them and going, gee is my access going to start getting restricted, you know will I have to pay a lot more money, will I have to add a zero to my to my monthly subscription, what what's going on. And then of course you've got this financing round which is less cash and more infrastructure and perhaps it's a way to kind of help the company kind of limp towards IPO so there's liquidity event so the you know original investor can kind of make a bit of a return.","offset":2574,"duration":28},{"text":"Matt: It turns out that you know this whole, it's pretty funny that this whole AI compute space is predicated on 600 billion dollars a year of CapEx which is incredibly energy-intensive when at the same time you're bombing the you know an area where 48% of the world's energy is and you know you're relying on energy being cheap in order to have this AI AI boom work.","offset":2602,"duration":24},{"text":"Eric: Let's move on to private credit. There's a lot that's been said already as a lot of private credit funds are gating investors that this is AI driven or it's related to Claude Code specifically creating fears that software companies would no longer be profitable. And I have a hard time with this one because this sounds like the claim I remember hearing in the 1970s when supposedly the introduction of the Hewlett Packard electronic calculator was supposedly going to put every accountant out of business and create vast unemployment of bookkeepers and so on and so forth and it was the exact opposite. It created more productivity. It seems to me that Claude Code just gave the software industry the biggest productivity boosting tool that they've ever had, and that's the reason that private credit is blowing up? Is that really right? Am I missing something?","offset":2626,"duration":54},{"text":"Matt: Well, I mean there's a few things going on. First of all, the these funding rounds are getting too big for equity, so as we saw with 122 billion and all the infrastructure, the vendor financing in there. And so you've increasingly these companies are turning to debt, even Meta had to go and get 30 billion from Blue Owl not so long ago in order to fund data centers because it's [44:59]","offset":2680,"duration":20},{"text":"Host: ...not do it off the balance sheet anymore and the numbers are starting to get too big for equity raising, so they're borrowing the money. And not just are we seeing record equity rounds, we're seeing record debt rounds. That BlueOwl financing of Meta was the biggest private credit round ever.","offset":2700,"duration":25},{"text":"Host: The problem is that private credit in these portfolios has got SaaS businesses and one of the big pitches that these AI compute companies has is that SaaS is dead, which is kind of ironic when ChatGPT on a $20 a month subscription is a SaaS business.","offset":2725,"duration":22},{"text":"Host: So it's kind of funny that they're saying SaaS is dead when they actually are SaaS businesses themselves. But there's been some warnings that have come out around these private credit portfolios that pack in the AI debt as well as the SaaS debt, because AI is actively pitching the future of these SaaS companies will be eroded by the fact that AI is coming.","offset":2747,"duration":26},{"text":"Host: So it's causing some instability in the private debt markets. And of course, what else is causing instability is rising interest rates in an uncertain world and a war in Iran amongst other things.","offset":2773,"duration":18},{"text":"Host: Matt, speaking of data centers blowing up, let's talk about the Iran conflict, its connection to AI, and particularly in your latest missive, you express some concerns that there's a risk that potentially this Iran conflict kind of pulls the rug on the whole AI business model. What do you mean, what's going on?","offset":2791,"duration":29},{"text":"Matt Barrie: Well, this is where the Fifth Industrial Revolution meets the Islamic Revolution, right? Um, a lot of the financing for AI has come from the Middle East. And you can imagine now if you're Saudi Arabia or you're the UAE and you have a fiduciary duty to protect your nation and your citizens and your economy, are you going to be putting it into hyper-rounds of Sam Altman's highly inflated valuations, or will you spend it on defense, energy, rebuilding your civilian and industrial infrastructure and potentially going to war to fund an army?","offset":2820,"duration":45},{"text":"Matt Barrie: You know, it's quite problematic when you've got a country that sits right in the middle of, you know, 48 percent of the world's energy infrastructure, that controls a strait where 21 million barrels of oil goes through every day, that can fire $20,000 drones at scale into anything within a 2,000 kilometer range and migrate not just your data centers in the region, but potentially your energy infrastructure into the cloud, literally in a puff of smoke.","offset":2865,"duration":37},{"text":"Host: Matt, another theme that you have in your latest missive is that you've got to be pretty smart to really get the most out of AI. What do you mean by that and what are the consequences?","offset":2902,"duration":16},{"text":"Matt Barrie: When I look at, deep down, at kind of what what's happening in the space, I mean, despite you've got a conflation of few things happening right now. You've got the AI companies trying to justify these huge valuation rounds and they're doing that by saying they're going to take away a lot of the world's work. Then at the same time, you've got sort of mass layoffs happening in the market with Block today, with Oracle, I don't know if you saw overnight, but Oracle has announced 18 percent of their workforce has been cut, etc.","offset":2918,"duration":37},{"text":"Matt Barrie: And so you would not be surprised that the general market has kind of conflated these things and thinking that AI is taking people's jobs away. Now, I see AI instead as a tool. It's a very, very powerful tool. It's, you know, but it's a productivity tool, much like the world was, you know, when you went to work and there was no computer on your desk and then you went to work and there was a computer at your desk or you didn't have mobile phones and you have mobile phones or you didn't have the internet and then you have the internet. Yes, there is incredibly disruptive and transformative time and some jobs are lost, but just like pretty much all forms of technology, more jobs are created over time.","offset":2955,"duration":51},{"text":"Matt Barrie: And what I mean essentially by you've got to be smart to use AI is, you know, AI is a power tool and so is a chainsaw, right? You give a chainsaw to a carpenter and they can do, a skilled carpenter, and they can do amazing things. You give a chainsaw to a novice and you can cause all sorts of problems, right? And so what I am observing, both across my platform as well as within my engineering team, is the people who are benefiting most from AI are the ones that are highly skilled and intelligent and know how to use the AI.","offset":3006,"duration":43},{"text":"Matt Barrie: And they're seeing productivity gains which are astronomical. They're seeing, you know, double to triple their productivity and you can measure that in different ways in terms of what they're achieving. And then as you go down the skill level, you do, you know, it is a rising tide lifts all boats. So if you're an average copywriter, you can now be a good copywriter. If you're an average illustrator, you can be a good illustrator using these various tools. If you, I make a joke, if you're an average programmer, well, you certainly are a confident programmer now using these tools. But to really get the best out of them, the people who are highly skilled are the ones that are really, really driving the outcomes. And I see that for example with your work and what you do, you writing these missives and books and so forth. And so I think just like technology has over time created a bifurcation in society where you have the people who are skilled and can create technology and that's where all the wealth flows.","offset":3049,"duration":68},{"text":"Matt Barrie: And that's why you see these companies with huge valuations and, you know, the ultra-high net worths in the technology industry, and then you see a general deterioration at the lower end of society. I think this is going to drive even further to an extent. People who really, really know how to use AI well are going to capture incredible opportunities in all sorts of different market segments where they can control that customer interface.","offset":3117,"duration":30},{"text":"Host: Well, I definitely agree with you. I just finished a writing project, or I should say Claude and I just finished a writing project that was considerably bigger than writing my book Beyond Blockchain. It required a lot more research because Beyond Blockchain was just my own opinions. It was a lot to it. And Beyond Blockchain took me more than three months, this project took me less than two weeks. And it's just amazing how much you can accomplish, although you definitely, as you say, it takes some skill to learn how to manage context window exhaustion and recognize symptoms of it occurring and so forth. So I couldn't agree with you more.","offset":3147,"duration":51},{"text":"Host: It kind of scares me though, Matt, because I think one of the biggest problems that society faces in the mid-2020s is the K-shaped economy. The tendency that the rich get richer and the poor get poorer. It sounds to me like AI really is going to exacerbate that in the sense that the smartest people are going to really benefit in productivity from AI. They're going to be a whole lot smarter and more capable than they used to be. One guy will be able to do the job of 10 or 12 guys. But the 10 or 12 guys that weren't that bright and really didn't figure out how to recognize the symptoms of context window exhaustion in their LLM, I think it really does take their jobs. And it seems to me like this could become the basis for deeper division in society.","offset":3198,"duration":55},{"text":"Host: So, Matt, I want to ask you this, because as the CEO of Freelancer.com, and for any listeners who aren't familiar with Freelancer, it's basically a marketplace where people can hire independent workers, whether they be high-end expert consultants in their field who are the leaders of their field, or if it's just a guy who'll design a logo for you for five bucks. You can hire all those different people on Freelancer. Matt has a very unique perspective on AI because first of all, he's running a company that uses AI, so he leads an engineering team that's building AI-based solutions. But some of the people that his customers are hiring are AI experts that are helping companies to implement AI at a corporate level. And then some of the more mass freelancer, larger group of graphic designers and people that are making logos and so forth, are users of AI that are leveraging their logo design contests and all that. So Matt sees all these different dimensions of different people using AI in different ways, corporations and organizations adopting it, versus an individual guy in Indonesia who's just trying to make a buck as a graphic designer suddenly becoming much more productive and being able to work in different languages that he doesn't even speak. So Matt sees all these different dimensions. Matt, from that vantage point that you have, what would you say are the top three things that you've become aware of that the average person who can't see all that stuff probably wouldn't think of?","offset":3253,"duration":122},{"text":"Matt Barrie: Well, the amazing thing is we're going to see the ability for you to get things done that you could never possibly think of before, right? Rather than just getting a website built, you'd get a whole business built. The ability for, you know, I think we're going to enter a whole new world of explosive entrepreneurship where now you really just need to have an idea to start a company and the ability for you to be able to execute on that using both AI and also access to humans powered by AI will be unprecedented. You'll be able to do it at a cost that's cheaper than ever before. So we're going to enter, I think, into an explosion period of hyper-competition and, you know, thanks to the internet and the ability to distribute products or services over the internet at scale so quickly, if you've got a great idea, that great idea can take off and you can make a billion dollars faster than any time ever in the history before.","offset":3375,"duration":57},{"text":"Matt Barrie: So I think it is an incredible time. I certainly haven't seen as yet the complete replacement of someone in a job. I don't know, you know, if you ask around, does anyone know any graphic designers that completely lost their job? Where the dislocation will occur is where you've got highly paralyzable workflows where you've got thousands of people or hundreds of people doing the same job. So maybe in a call center, where there might be 10,000 people or 1,000 people in a call center doing the same workflow which is customer support, answering a phones, etc. With AI, you'll be able to take it from 1,000 people down to maybe 100 people. You know, if you've got 30 junior lawyers drafting legal agreements in a room, maybe you'll be able to take it down to 13 people in a room.","offset":3432,"duration":53},{"text":"Matt Barrie: But I don't see, because of the way AI works and goes in circles and the failure mode and the fact that as you burn more tokens you've got more chance, even though the unit rate of errors goes down, the fact you're burning all these extra tokens means that ultimately, you know, an error becomes more catastrophic in a way that a human never would make an error. What that means is that I think the ultimate combination is humans and AI. I think it's probably one of the greatest productivity tools known to man and I think it's going to open up a whole amazing golden age of building and creating businesses and hyper-competition.","offset":3485,"duration":46},{"text":"Host: Well, Matt, I can't thank you enough for a terrific interview. Before I let you go, you run Freelancer.com, a public company that trades under ticker symbol FLN on the Australian Stock Exchange. Tell people for who are not familiar with it what Freelancer does and how to follow your work. Your latest piece, again, it's linked in the research roundup email, it's published on Medium, it's called Pay to Pray. You write quite a bit of interesting stuff for people who want to follow your work and learn more about Freelancer.com, tell us about it.","offset":3531,"duration":49},{"text":"Matt Barrie: Well, we run the world's largest cloud workforce, so it's about 87 million people in that marketplace. We can do any job you can possibly think of from $10 jobs to $10 million jobs. You know, the mainstream jobs are things like build me a website or build me an app or, basically help me start a company or grow my company. The biggest things we run, we've got a moonshot innovation challenge program at the high end where we help all sorts of US government departments and enterprises around the world solve scientific and technological challenges. So, you know, give us your hardest scientific or technological challenge and we'll solve it or you don't have to pay the prize out. So the biggest thing we've got running right now is the seven and a half million US dollar gene editing challenge for the central nervous system of humans, which we're doing for the National Institute of Health and also working with NASA.","offset":3580,"duration":66},{"text":"Matt Barrie: Actually, ironically, when Artemis goes up in the next 24 hours into space, we worked with NASA to crowdsource the mascot from kids all around the world that's going up with the astronauts to basically inspire kids about space and so forth. If you want to get anything done, you come to our site, we can get it done, and no job is too small or too big or too complex and it ranges through to mechanical engineering or electronic design or whatever you need. And if you want to follow my writing on AI, there's a series of, obviously a podcast that we've done together on MacroVoices on AI, I think this is maybe the fifth or the sixth. And if you go to Medium or Substack you'll be able to find me and see my essays.","offset":3646,"duration":59},{"text":"Host: And if you just put Matt's name in, Matt Barrie, at the search bar at macrovoices.com, you'll see a list of all of the previous AI interviews that we've done right here on MacroVoices. Patrick Ceresna and I will be back as MacroVoices continues right here at macrovoices.com.","offset":3705,"duration":26},{"text":"Host: Eric, it was great to have Matt back on the show. Now Dr. Anas Alhajji is next on deck for a special second feature interview on the developing Iran conflict and what it means for the oil markets. Then Eric and I will be back for our usual postgame chart deck and trader of the week. Since this extra coverage format seems to be a hit with listeners, we'll do our best to continue it for as long as the situation in the Middle East warrants. Now let's go right to Eric's interview with Anas.","offset":3731,"duration":63},{"text":"Host: Joining me now is Energy Outlook Advisors founder and managing partner Dr. Anas Alhajji. Anas, it's been three weeks since we had you on the program for the first Iran war update. It's been one month since this conflict started. President Trump just moments before we began recording this segment gave an address to the American people. Let's start with the big picture update since this began. How has everything evolved, how have your views evolved since we had you on three weeks ago, and what did President Trump need to achieve in this address and did he achieve that?","offset":3794,"duration":43},{"text":"Dr. Anas Alhajji: Well generally speaking, we got to frame what has been going on and President Trump basically did not frame it. So, either this is about Iran and its nuclear and he tried to make it that way, or Iran part of a bigger picture and the audience basically can choose one of those two. So either it is about Iran and its nuclear program and the outcome of that basically and its impact on the market are completely different from the idea that Iran is part of a big picture. If you look at trade wars, tariffs, sanctions, Venezuela, Panama, Red Sea, China, Greenland, etc., so is Iran and this war is part of it or Iran is completely separate. The president today basically tried to make it about Iran, but all of a sudden he deviated from that and he started talking about we don't need the Middle East, we don't need oil from there, buy oil from me. And this makes the second option basically kind of more credible in this case.","offset":3837,"duration":70},{"text":"Dr. Anas Alhajji: Anyway, we'll go back to that later on. And the other issue is, who closed the Hormuz Strait? You have only two answers. It's either the United States through the insurance companies or Iran. And it's up to you, the audience, basically to believe one of them. There are evidence on both sides. But one of the ironies here is that when we talk about the details of who closed the Hormuz Strait and we say Iran has no power to close it, then people say, well this is a conspiracy theory. It turns out it's a bigger conspiracy theory to say Iran closed the strait. And we heard the president today saying that Iran's navy is gone. Literally said that. And he said they have no radars, he talked about the total devastation of course their leadership is gone, all that stuff. So to believe that we have all these attacks and the aftermath of this devastation that happened to Iran.","offset":3907,"duration":64},{"text":"Dr. Anas Alhajji: We have 28 countries are in charge of protecting the Gulf and the Hormuz Strait through an organization that is based in Bahrain. You have the US navy, you have the Indian navy, you have the British navy, you have the French navy, you have the Saudi navy, you have all the other countries' navies in the area. And then after all this devastation Iran still control the Hormuz Strait? This goes against everything that President Trump said today. So the story does not match in this case. So to sum up, either you look at this war as it's either Iran and nuclear or it is part of a bigger picture of the changes going in the world and then you look at Hormuz Strait and say well it's either the United States and the insurance companies closed it and the main beneficiary of this is the United States and we heard President Trump talking about it today. Or Iran closed it but that does not make any sense and that is a bigger conspiracy theory because how was Iran able to do that after what we heard from President Trump today and the existence of all the other navies.","offset":3971,"duration":64},{"text":"Dr. Anas Alhajji: Unfortunately President Trump today failed to deliver to the nation the message why the war is still going. Many people basically expected that he will announce some sort of an end to the war. We told our clients couple of weeks ago that all the indications point to a long war, and today we got the confirmation. This is not going to end soon. And if it's not going to end soon, there is another question mark about what President Trump said. If everything is destroyed in Iran, then why is this war continuing? What is left to be destroyed? So the impression I got out of this speech is that the people around President Trump are not telling him the truth and he is living in his own ivory tower without looking at what's going on in reality. And we have countries around the world that are literally having blackouts, country do not have propane to cook, countries basically are struggling to get petroleum products, etc. Many countries are worried about food supplies and the president did not even mention it today. This is a global crisis. This is the largest global crisis we have in our lifetime. And the president even did not talk about it and he did not explain to the American people why this war is still going until now. If everything is destroyed why the war is not ending?","offset":4035,"duration":93},{"text":"Host: Anas, who is actually in charge in Iran and calling the shots? Because we've seen so much conflicting news in the last few days. President Trump before this speech had said very clearly in his Truth Social posts that what the speech was likely to be about is that Iran had requested a ceasefire and that Trump was likely to agree to that ceasefire. Now preempt, apparently trying to preempt this speech, the guy who is still technically the president of Iran, although the president is not really the head of state who's in charge in Iran, it's really the Supreme Leader who calls the shots. But this guy who is the president issued a letter to the American people in which he says look we don't want any ceasefire, we would not agree to a ceasefire, we want the war to be completely over and we won't accept any end until we have a promise of no further attacks.","offset":4128,"duration":59},{"text":"Host: So it seems like if nothing else that whether that was a true statement that's reflecting the Iranian government or just the guy who's not really in power anymore expressing his view isn't clear, but it seems like it did have the effect of taking the ceasefire discussion out of President Trump's speech. Meanwhile, we're told that who's really in charge of Iran is the Supreme Leader, but that would now be the killed Supreme Leader Ayatollah Khamenei's son. Nobody's seen him since he supposedly took power. Nobody's heard from him. We're told that he's still alive, but we don't know that for sure. Who's calling the shots in Iran and how should we interpret the various messages that we're hearing out of the media?","offset":4187,"duration":53},{"text":"Dr. Anas Alhajji: The reason why we call it the Iranian regime while in other places we call it government, but in Iran we call it a regime, because we have the official government and then we have the support coming from various militias and groups of that government. If you sum them up we call that a regime. So the Iranian revolutionary guard are not part of the government, but once you put them together they become a regime. And what we've seen historically is that there are of course differences between the government itself, the official government where you have prime minister and you have president and you have a foreign minister, they are more diplomatic. While most of the military actions and the attacks and everything else comes from the revolutionary guards who are not part of the government. So historically we've seen that division.","offset":4240,"duration":56},{"text":"Dr. Anas Alhajji: But even within the revolutionary guard they are not really one unit. And what happened is that after the killing of most of the leadership including Ayatollah Khamenei who the religious leader, they decided to at least for a short period of time to work independently. So they told them okay we have no leadership at this stage so every group you are on your own. So we've seen various groups basically acting on their own, attacking on their own, based on their location, based on their beliefs, based on whatever their leaders basically believe about other countries. So we've seen attacks on the Gulf nations, we've seen more attack on some relative to the others, etc. And then the foreign minister came in and said we know nothing about it. Probably he's right because there was a complete disintegration of the system at least for that period of time.","offset":4296,"duration":65},{"text":"Dr. Anas Alhajji: At this stage, it's very clear that the revolutionary guard basically took over the government. So now we can say that unlike in the past where we have a government and then we have the revolutionary guard and the others, now the revolutionary guard basically literally took over the government, so they are one unit right now. And the whatever kind of nice messages we get, we get from people who have no actual power on the ground. And so everything you hear about the Iranian parliament, this is just a joke. This has no impact. So when they say oh we are going to install a tollbooth for example and we got to charge $2 million per ship, etc., this is coming from people who have no power on the ground at all. Unfortunately the western media basically took them seriously and spread the misinformation all over the world. But they have no power.","offset":4361,"duration":62},{"text":"Dr. Anas Alhajji: And speaking about Iran's ability to charge for it, under international law there is no way that they can charge. There is no way they can charge. And whatever they are doing if they are getting any money from anyone, this is an extortion. And the funny part here is if there are governments that are literally cooperating with Iran and paying them so their ships can pass, so we are talking here about Pakistan, we are talking about India, we're talking about Malaysia and probably others, why President Trump is not mad at them? Because they are cooperating with those terrorist and the regime that is killing people who killed 40,000 people of its own people. Why? Because if they are paying them money then they are supporting the regime. Why we did not hear a word about that? Simply because probably it does not exist in the first place.","offset":4423,"duration":59},{"text":"Host: Anas, I want to follow up on a really important point that you made in your last interview three weeks ago. What you explained is that most people assume that Israel has nuclear weapons that they could use against Iran but Iran doesn't have a nuclear weapon that they could use in retaliation. You explained that Iran really does have the equivalent of a nuclear weapon in terms of the consequence of using it, which is that Iran could target the desalination facilities of both Israel and other allied Gulf countries which unlike Iran itself which only gets 3 percent of its drinking water from desalination, Israel and other countries in the region are extremely dependent on desalination. So if Iran attacks their desalination they are forcing starvation and just horrible, horrible humanitarian outcomes. They don't have that same vulnerability with their own desalination plants because they only get 3 percent of their own water from desalination. And so you made the argument that Iran could potentially if they really wanted to use the proverbial nuclear option they could escalate to targeting desalination facilities.","offset":4482,"duration":79},{"text":"Host: Now about a week after you made that prediction, Iran made a formal announcement saying if our energy, our civilian energy infrastructure is targeted, we will respond by targeting the enemy's desalination facilities which could bring about that horrific humanitarian outcome that you described. So it seems like this is escalating and escalating quickly. Do I have that right and how do you see this playing out? How serious of a risk is this?","offset":4561,"duration":35},{"text":"Dr. Anas Alhajji: Yes, you are absolutely right. And one of the issues that we have to really realize is when we say this is a historic event, it's not only in term of impact. It is historic in various ways. One of them is there are no red lines. There are no, no one basically has any red lines, the United States, Israel and the Iranians. No red lines. And if you look at the attacks of the other countries in the Gulf etc. there were no red lines. So on every part on the parts of everyone there are no red lines and that make it more, in a sense, another reason to be historic. But one of the other reasons why it is historic because of the nonsense, there are a lot of nonsense going on that we cannot even explain at this stage. We don't know what's going on. For example, Iran is hitting the GCC countries, that's the six countries in the Gulf and some targets in Iraq.","offset":4596,"duration":57},{"text":"Dr. Anas Alhajji: They are hitting them because they said oh you are cooperating with the Americans and the Israelis and therefore we got to hit you. Well if that's the logic then we need to explain this. The Israeli planes that are hitting the targets in Iran and the Israeli rockets basically that are using the fuel, that fuel is some of it is coming from Azerbaijan and Azerbaijan is next to Iran. It's a neighbor. And Chevron is the largest oil company operating there in an oil field. So it is an American company. We saw Iran hitting for example a Ruwais refinery in the UAE. Ruwais refinery is a joint venture between ADNOC and European companies, so there are no American presence there and there are no military bases there. Yet they attacked it.","offset":4653,"duration":55},{"text":"Dr. Anas Alhajji: But why they are not attacking Azerbaijan who is literally giving the oil to Israel to bomb targets in Iran? Same thing for Kazakhstan, the same thing for Turkey because that oil coming from Azerbaijan basically is coming through Turkey. People saying well Turkey is a NATO member. Well I can tell you we know that Trump does not care about NATO and Trump does not care about Turkey if Iran attacks Turkey. But there are, all this confusion and all those issues that do not make sense at all. For example, attacking Salalah oil depots in Oman. Salalah is like, if you are in Iran Salalah is at the other side of the world. And it's an oil depot. Why you are hitting it? There are no Americans there, there are no Israelis there, there's no military base there. And how come literally a drone will go all the way across the Emirates, across Saudi Arabia, across the empty quarter, go all the way to the end of Oman to the other side where Yemen is and hit the target precisely?","offset":4708,"duration":67},{"text":"Dr. Anas Alhajji: So there are too many questions, there are too many things that do not make, do not make sense. For example, hitting this refinery. If Hormuz Strait is closed and the refinery cannot export, why you are wasting ammunition, why you are wasting drones or rockets to hit it while the refinery basically is useless? Why hitting a tanker like the Kuwaiti tanker yesterday that's been in the same place for a month? It's 200 mile away from Hormuz Strait. It cannot go away, it cannot go through the Hormuz Strait, is not even moving. Why hitting it while you know it's not going to go anywhere? And was it a mistake, was it, was the objective something else? What, whatever the case is there are too many questions that have no, no answers. But the bottom line here is the Hormuz Strait is still closed and that is a big failure for the US policy in the region regardless of what Trump claimed earlier in the speech.","offset":4775,"duration":69},{"text":"Dr. Anas Alhajji: The fact that the Hormuz Strait is closed, even at this moment, regardless of who closed it, is a failure of US policy. And there are strange things again when we talk about historic event, there are some very strange things. I'm going to tell you some. For example, if we go back to the national security strategy, it was released in November about four months before the war. So I'm going to read this to you from that strategy. So I'm reading right now. \"America will always have core interest in ensuring that Gulf energy supplies do not fall into the hands of an outright enemy. That the strait of Hormuz remain open, that the Red Sea remain navigable, that the region not be an incubator or exporter of terror against American interest or the American homeland, and that Israel remain secure.\"","offset":4844,"duration":62},{"text":"Dr. Anas Alhajji: Let's look at this one because this is kind of really strange statement. President Trump today said we don't need the Hormuz Strait, we don't need the Gulf, we don't need any of that. The strategy said we don't want it to fall into enemies' hand. And he did not say that today. This was one of the biggest failures of the speech. He should have said that. But here is the strangest statement in the strategy. Of course the strategy is really conclusions of a larger document that remains secret. We don't know what's in it. And therefore this, this paragraph, several paragraphs that been omitted. And here is why I'm saying this. The United States historically if you look at various strategies over the last 50 years, always talked about freedom of navigation in those waterways or the choke points. Notice the following.","offset":4906,"duration":54},{"text":"Dr. Anas Alhajji: This is the first time we are, we see a document like this, four months before the closure of Hormuz Strait, saying \"The Strait of Hormuz remain open.\" It never been closed. This statement makes sense if the Hormuz Strait was closed just once in history. So whoever wrote that statement there they must have something before this paragraph, \"open.\" If it wasn't closed in history, why you are using this statement, \"remain open\"? And the other thing is you are talking about Hormuz Strait, why you are tying Israel to it? You look at what we have in actual life, we have Israel basically attacking Iran right now and the Hormuz Strait basically is part of it. But the question here is if Anas Alhajji wrote in June and July, that months away, months before the war, months before the closure of Hormuz Strait.","offset":4960,"duration":61},{"text":"Dr. Anas Alhajji: And people who they can go to our daily energy report and they can go to our newsletter and they can see it. I wrote the 12-day war. In June Israel attacked Iran. China had no problem. Everyone was expecting Israel to attack Iran for the last 20 years and finally they did. But once the Trump administration started attacking Iran, that changed everything. And China was ahead, they knew in advance what's going on. So China basically stopped importing LNG from the United States, stopped importing oil from the United States, they were ahead on Venezuela before the blockade, they were ahead on Venezuela before the death of Maduro, etc. But they been and if you go back to the previous shows we had probably the last three shows or four, we've been talking about how China was building those massive inventories, oil inventories and everything else, in preparation for either war, sanctions or a major interruption.","offset":5021,"duration":62},{"text":"Dr. Anas Alhajji: Now we know. China knew. And China one of the least impacted countries by the closure of the Hormuz Strait right now at least in the short run. Once we look at those events and look at the 12-day war, what I've written was the China got the message out of the stories that been published in major news media outlets and very prestigious outlets. We've seen a number of stories talking about Iran closing the Hormuz Strait. Again we are talking about June. And if you look at those stories you will find something kind of very strange about them because all of them have the same talking point, some of them in the same sequence. That means there is a public relation company basically was directing that. Why they were doing it? So what I wrote at that time was China got the message and the message was the Hormuz Strait will be closed, but it's not Iran who is going to close the Hormuz Strait, it is the United States.","offset":5083,"duration":65},{"text":"Dr. Anas Alhajji: Again this is written in June and July. So we have the November document after that talking about the keeping the strait open and the strait never been closed. And then in June and July we were talking about the United States basically closing the Hormuz Strait. And then we have the insurance fiasco that we've seen until now. I know that many people were pushing against this, this idea, calling it a conspiracy theory etc. and they tried really hard to debunk this idea. Here is the issue that they need really to answer. The issue is this. Even if you send the navy to escort ships out of the Hormuz, ships are not going to move without insurance. If the insurance remains non-existent or extremely expensive, ships are not going to go along with the navy. Because shippers are going to tell Trump or Modi or anyone else who wants to send those navy vessels, telling them look, if I go with you and I get attacked and I lose my ship and I lose everything on it, will you compensate me? And Trump is going to tell them no, Modi is going to tell them no, I'm just providing a service for you you should pay me for it I don't have to compensate.","offset":5148,"duration":73},{"text":"Dr. Anas Alhajji: They are not going to go. They need that insurance to exist and to be cheap enough for them to move. So any way you look at it you say it's conspiracy theory or not it doesn't matter. They need that insurance. So it is an insurance story anyway you look at it. On the other side those who are claiming that Iran closed the Hormuz Strait again Trump told us today that Iran has no navy, has no power, Iran, we heard the whole talk. So either someone is lying to President Trump or Iran has no power.","offset":5221,"duration":39},{"text":"Host: Anas, there's one more nuclear option that we need to talk about which is so far in human history, no military force has ever intentionally targeted an operating nuclear reactor. In other words the horror scenario, the terrorism scenario of intentionally dropping a bunker buster bomb on top of an operating nuclear reactor to turn it into a dirty bomb of sorts. Now one of the things that has already occurred, and by the way article 56 of the Geneva Convention very clearly prohibits the intentional targeting of any nuclear electrical generation station. You cannot target nuclear electrical generation stations by international law. Except either the United States or Israel, I'm not sure who, has already made at least three attacks, air strikes on the Bushehr nuclear power plant which is Iran's one operating nuclear power plant.","offset":5260,"duration":63},{"text":"Host: Now those were not targeting the reactor. If they had intended to blow up the reactor and breach its containment they would have succeeded, so they weren't trying to do that. They were probably trying to take out the infrastructure and prevent it from generating electricity and supplying it to the grid. But it does seem that they're already violating the Geneva Conventions by targeting a nuclear power plant. Where is this headed? I'm just really concerned about the escalation that could occur from here.","offset":5323,"duration":33},{"text":"Dr. Anas Alhajji: Again this is a historic event for several reasons and one of the reasons why because there are no red lines. And what you described is a red line that's been crossed and is going to be crossed again and again and as a result Iran is going to cross this red line on the other side. Because as you know they already listed the four reactors in the UAE as a target, the stations like you said probably not the reactors because they need kind of special bomb busters or others etc. But they literally put that as a target. So there are no red lines and with a war with no red lines everything is possible.","offset":5356,"duration":42},{"text":"Dr. Anas Alhajji: The issue that I experienced firsthand was when I woke up in the morning here in Dallas and I got flood of messages and emails, people scared to death from oh what if Bushehr basically explodes and would we get nuclear and literally some high net worth individuals basically sent me messages and said should I put my family in a car and leave to Oman or any other country. So people basically being panic and that panic alone could cause massive problem because if you are talking about millions of people taking the highways, the rumor on its own is enough to cause problems let alone if it, if it becomes reality.","offset":5398,"duration":48},{"text":"Host: Well as you said Anas, Iran has announced that if their civilian power generation infrastructure is targeted, they will respond number one by targeting the desalination facilities of neighboring countries which you've described as a nuclear-like option because of its consequences. Number two that they will target the United Arab Emirates operating nuclear power station at Barakah, or the Barakah nuclear power station which is outside of Abu Dhabi. They've already said that will be a target. And then in tonight's, this is Wednesday night that we're recording, in tonight's address from President Trump, he said very soon and possibly simultaneously that the United States might target all of Iran's civilian power generation stations. So it sounds like the event that Iran said they would respond to by targeting both desalination and nuclear power plants is one that President Trump just signaled his intention to escalate to. This is a very scary moment. This feels like, you know, a replay of the Cuban Missile Crisis except this time we actually know about it. Am I exaggerating to say it's that big of a deal?","offset":5446,"duration":72},{"text":"Dr. Anas Alhajji: No, but I think we should be very aware of the fact that today after Trump speech, I am completely convinced that the people around him are not telling him the right picture. The way he described things are far away from reality. Let me give you another example on how far away from trade reality he said drill baby drill. Okay, drilling activities in the United States been down since he came to office. US production been declining in recent months. So he is completely out of reality and I am afraid that the people around him are not telling him the truth. Before this speech oil prices were down. Right now after the speech oil prices are 5 percent higher and they are going up. So the message, the main message of the speech basically is that this is a long war and this is going to continue and the market is pricing it.","offset":5518,"duration":57},{"text":"Dr. Anas Alhajji: And I think by tomorrow morning probably many people who were counting on probably it will be a matter of days is going to change their mind. And that will impact of course the markets whether you talk about oil gas LNG or anything else.","offset":5575,"duration":14},{"text":"Host: Anas let's move on now to what everyone's waiting for, which is how do we bring this all together in terms of the outlook for oil prices, LNG, energy markets generally. If the US is basically from what President Trump said very directly in tonight's public address, he said the US will probably walk away from the situation in the Hormuz Strait, leave it to other countries that depend on it. It sounds again as you suggested that maybe President Trump isn't getting the full story from the people around him because of course oil prices are set globally. If the Hormuz Strait is not reopened, energy prices are going to stay high for everyone, including US voters through the midterm elections. So it seems like the story doesn't add up.","offset":5589,"duration":55},{"text":"Host: But it does seem like the president intends to walk away from the situation in the Hormuz Strait once they've finished their military operations against Iran and it's unspecified exactly what that is, but the one thing he did say that was a specific indication of intentions was to target potentially simultaneously all of Iran's civilian energy production capabilities, exactly the thing that Iran has said if it happens is what's going to cause them to target desalination and nuclear power plants. What does this mean for energy prices and what does this mean for global energy markets more broadly?","offset":5644,"duration":36},{"text":"Dr. Anas Alhajji: When President Trump is talking about others basically coming and protecting the Gulf and opening the strait, he means only the people he know and he is talking about European leaders. He's not talking about China. He's not talking about the rest of Asia. So let's be clear about that. China will be happy to bring its navy and and have control of the strait. This will never happen and the United States will...","offset":5680,"duration":25},{"text":"Host: The audio cuts there.","offset":5705,"duration":2395},{"text":"...or sponsors. MacroVoices, its producers, sponsors, and hosts Eric Townsend and Patrick Ceresna shall not be liable for losses resulting from investment decisions based on information or viewpoints presented on MacroVoices.","offset":8100,"duration":15},{"text":"MacroVoices is made possible by sponsorship from bigpicturetrading.com and by funding from Fourth Turning Capital Management LLC. For more information, visit macrovoices.com.","offset":8115,"duration":15},{"text":"[Intro Music]","offset":8130,"duration":15},{"text":"[Intro Music continues]","offset":8145,"duration":15},{"text":"[Intro Music fades out]","offset":8160,"duration":15}],"logs":[{"elapsed":"0.0","message":"Downloading audio from YouTube...","detail":null},{"elapsed":"0.0","message":"Trying download with browser cookies (ad-free)...","detail":null},{"elapsed":"2.8","message":"⚠ Cookie download failed: WARNING: [youtube] [jsc] Error solving 2 challenge requests using \"deno\" provider: Error running deno process (returncode: 1): \u001b[0m\u001b[1m\u001b[31merror\u001b[0m: Uncaught (in promise) TypeError: Cannot read prop","detail":null},{"elapsed":"2.8","message":"Retrying without cookies...","detail":null},{"elapsed":"68.0","message":"⚠ Downloaded without cookies — audio may contain ads","detail":null},{"elapsed":"68.0","message":"Audio downloaded (74.9 MB) in 68.0s","detail":"File size: 74.9 MB"},{"elapsed":"68.0","message":"Video title: MacroVoices #526 Matt Barrie: Pay To PrAI","detail":null},{"elapsed":"68.1","message":"Audio duration: 2:15:41 (135.7 min)","detail":null},{"elapsed":"68.1","message":"Large audio (135.7 min) — will use chunked transcription with gemini-3-flash-preview","detail":null},{"elapsed":"68.1","message":"Skipping full-file attempt — using chunked transcription for 135.7 min audio","detail":null},{"elapsed":"69.3","message":"Split audio into 4 chunks for transcription","detail":null},{"elapsed":"69.3","message":"Transcribing chunk 1/4 (starts at 0:00)...","detail":null},{"elapsed":"69.3","message":"Uploading audio to Gemini File API...","detail":null},{"elapsed":"73.0","message":"Audio uploaded in 3.7s","detail":"File ref: files/4lxpyom0ebzz"},{"elapsed":"73.0","message":"Audio processed in 0.0s. Transcribing with gemini-3-flash-preview...","detail":null},{"elapsed":"133.7","message":"Chunk 1: 75 segments, last timestamp 44:40","detail":null},{"elapsed":"133.7","message":"Transcribing chunk 2/4 (starts at 45:00)...","detail":null},{"elapsed":"133.7","message":"Uploading audio (offset 45:00) to Gemini File API...","detail":null},{"elapsed":"136.9","message":"Audio uploaded in 3.2s","detail":"File ref: files/a0pg2gieccn7"},{"elapsed":"136.9","message":"Audio processed in 0.0s. Transcribing with gemini-3-flash-preview...","detail":null},{"elapsed":"188.0","message":"Chunk 2: 59 segments, last timestamp 1:35:05","detail":null},{"elapsed":"188.0","message":"Transcribing chunk 3/4 (starts at 1:30:00)...","detail":null},{"elapsed":"188.0","message":"Uploading audio (offset 1:30:00) to Gemini File API...","detail":null},{"elapsed":"192.8","message":"Audio uploaded in 4.8s","detail":"File ref: files/n381uljlnyfr"},{"elapsed":"192.8","message":"Audio processed in 0.0s. Transcribing with gemini-3-flash-preview...","detail":null},{"elapsed":"245.2","message":"⚠ Empty response (attempt 1/3) — finishReason: RECITATION, blockReason: none","detail":null},{"elapsed":"245.2","message":"Waiting 10s before retry...","detail":null},{"elapsed":"307.9","message":"⚠ Empty response (attempt 2/3) — finishReason: RECITATION, blockReason: none","detail":null},{"elapsed":"307.9","message":"Waiting 20s before retry...","detail":null},{"elapsed":"375.5","message":"⚠ Empty response (attempt 3/3) — finishReason: RECITATION, blockReason: none","detail":null},{"elapsed":"376.0","message":"⚠ Chunk 3 returned empty response — skipping","detail":null},{"elapsed":"376.0","message":"Transcribing chunk 4/4 (starts at 2:15:00)...","detail":null},{"elapsed":"376.0","message":"Uploading audio (offset 2:15:00) to Gemini File API...","detail":null},{"elapsed":"377.4","message":"Audio uploaded in 1.4s","detail":"File ref: files/f5diboye4trt"},{"elapsed":"377.4","message":"Audio processed in 0.0s. Transcribing with gemini-3-flash-preview...","detail":null},{"elapsed":"380.8","message":"Chunk 4: 5 segments, last timestamp 2:16:00","detail":null},{"elapsed":"380.8","message":"Chunked transcription complete: 139 total segments","detail":null},{"elapsed":"380.8","message":"Total cost: 204,608 in / 21,281 out — cost: $0.1661","detail":null},{"elapsed":"380.8","message":"Total transcription time: 312.7s — 139 segments","detail":null},{"elapsed":"380.8","message":"Analyzing topics across 139 segments with gemini-3.1-pro-preview...","detail":null},{"elapsed":"435.0","message":"Topic analysis complete in 54.2s — found 19 topics","detail":null},{"elapsed":"435.0","message":"Analysis tokens: 22,053 in / 1,560 out / 5,828 thinking — cost: $0.1328","detail":null},{"elapsed":"435.0","message":"Pipeline finished in 435.0s — total cost: $0.2989 (255,330 tokens)","detail":null}]}