Latest / The Bear Market Podcast with Fexingo: Surviving Downturns, Buying the Dip, and Long-Term Resilience / The AI Energy Bill That Will Break Your Portfolio
Transcript
- Lucas: Jamie Dimon just dropped a number that should keep investors up at night: he expects hyperscaler artificial intelligence spending to hit one trillion dollars next year. That is not just a capital expenditure figure; it is a signal that the physical constraints of our power grid are about to collide with the theoretical limits of computing. Luna: One trillion in spending sounds abstract until you realize most of that money has to buy something tangible like servers, chips, and the massive amounts of electricity required to run them. Are we actually looking at an energy crisis disguised as a tech boom? Lucas: Exactly. The VIX is sitting at fourteen point two eight today, which suggests the market is pricing in calm, but that calm ignores the structural reality that data centers are consuming electricity at a rate that outpaces our ability to generate it. We have seen real GDP growth slow to one point five percent this quarter, yet the energy intensity of these facilities is doubling. Luna: So the disconnect is between macroeconomic growth and micro-level resource consumption. You are saying the economy might be stalling, but the appetite for digital infrastructure is accelerating regardless. Lucas: Precisely. Let us look at the ten-year Treasury yield, which ticked up to five point zero one percent yesterday. When borrowing costs remain elevated like this, companies cannot simply leverage their way into building new power plants or upgrading transmission lines. They have to find efficiency, or they have to pay a premium. Luna: And that premium is likely being passed down to consumers through higher electricity bills, or absorbed by tech giants who are quietly eating into their profit margins to keep the lights on. Lucas: Right. And here is where the opportunity lies for investors who are willing to look beyond the stock price of the chipmakers. The constraint is no longer silicon; it is electrons. We are entering a phase where the value of a company will be tied directly to its access to reliable, affordable power rather than just its code. Luna: That shifts the entire paradigm of how we evaluate tech valuations. If a cloud provider cannot secure enough power, their growth story hits a hard ceiling, regardless of how popular their AI models are. Lucas: Let us talk about the specific players. Microsoft, Amazon, and Alphabet are all racing to build out their respective clouds. But when you drill into their recent filings, you see a creeping increase in power purchase agreements and dedicated nuclear contracts. This is not marketing fluff; it is survival strategy. Luna: So we are seeing a vertical integration play where tech companies are effectively becoming energy utilities themselves. Is that a sustainable model given the regulatory hurdles involved in managing power grids? Lucas: It is messy, but it is happening. Look at the headline from Kalshi asking the CFTC to allow margin trading on their platform. It shows that financial innovation is trying to catch up to physical reality. Meanwhile, the actual physical assets—transformers, turbines, fuel—are facing supply chain delays that can stretch out to five years. Luna: Five years is an eternity in tech time. By the time those transformers arrive, the AI models running them will be completely different architectures. How do you plan for a bottleneck that moves slower than the innovation it supports? Lucas: You don't predict it; you hedge against it. Investors need to start looking at the utilities sector not as boring, stable dividend plays, but as high-growth infrastructure partners. The spread between short-term and long-term Treasury yields is narrowing, which usually signals recession fears, but in this context, it highlights the urgent need for immediate capital deployment in energy. Luna: I love that framing. We treat utilities like background noise, but they are actually the front line of the artificial intelligence war. If I am holding a portfolio heavy in software, I might be missing the hardware layer that makes it all possible. Lucas: That is a critical insight. Most retail investors are obsessed with the applications layer—the apps you use, the games you play. But if the underlying physics of cooling and power fails, the application layer evaporates. We saw glimpses of this during recent heatwaves where data centers had to throttle operations to prevent grid overload. Luna: Throttling operations? So the service quality actually degrades because there is literally not enough juice to go around. That must be terrifying for enterprise clients who rely on uptime guarantees. Lucas: It is. And that risk is being priced in very slowly. The S&P five hundred is up nearly three percent over the last week, driven largely by tech strength. But that strength is fragile if it depends on a power supply that is maxed out. We need to diversify into the enablers. Luna: Enablers like renewable energy firms, or traditional utilities pivoting to green hydrogen, or even companies specializing in advanced cooling technologies for server rooms. Lucas: All of the above. Advanced cooling is a particularly interesting niche. Liquid immersion cooling is becoming standard for high-density AI clusters, and the companies supplying those systems are seeing order books extend well into next year. It is a sub-sector that flies under the radar but is essential. Luna: So, liquid cooling isn't just an option anymore; it is a necessity for density. And since water resources are also under strain in many key data center locations, this becomes a multi-resource problem involving both energy and water. Lucas: Water is the other half of the equation. In arid regions where many data centers are located due to cheap land, using millions of gallons of water for cooling is becoming politically and environmentally untenable. Companies that solve this dry-cooling problem are sitting on gold mines. Luna: It is fascinating how a purely digital revolution is forcing us back into fundamental resource management. We are essentially re-industrializing the internet. Does this mean we should expect higher inflation as these infrastructure costs get baked into consumer prices? Lucas: Eventually, yes. Everything is eventually priced. But right now, the hyperscalers are absorbing the cost to maintain their competitive edge. Once the capex cycle turns and they need to show profits, those costs will trickle down. For now, we are in the investment phase. Luna: Which brings us back to the one trillion dollar figure. If that spending doesn't yield proportional revenue growth, we could see a massive correction in tech valuations. The market is currently rewarding ambition, but it may soon punish inefficiency. Lucas: Efficiency is the keyword. The winners will be the companies that can deliver more compute per watt. It is not just about having the fastest chip; it is about having the most efficient system. That is where the alpha lives. Luna: I think a lot of listeners are going to need a moment to process this shift from pure software metrics to physical infrastructure metrics. It changes the entire dashboard we use to monitor our holdings. Lucas: Absolutely. Start adding columns to your spreadsheet for power consumption and cooling requirements. It might seem tedious, but it will give you a clearer picture of the true cost of doing business in the age of artificial intelligence. Luna: Practical advice wrapped in a macro warning. I appreciate that balance. Before we dive deeper into specific sectors, let me take a quick breath. Lucas: If our conversations about navigating these complex market shifts have helped you clarify your own investment strategy, consider supporting the show. A couple of dollars a month is genuinely what keeps these independent investigations going. You can find us at buy me a coffee dot com slash fexingo. Luna: It really does make a difference in allowing us to dig into the details without rushing. Thanks for keeping the lights on here. Lucas: Back to the grid. Let us look at the nuclear angle. We are seeing a resurgence in interest in small modular reactors, specifically designed to power data centers directly. This bypasses the aging transmission infrastructure entirely. Luna: Direct connection? So instead of relying on the public grid which is strained, tech giants build their own mini-power plants on site. That seems like a huge capital risk but potentially massive long-term reward. Lucas: It is a bet on longevity. These reactors are designed to last forty years, which aligns perfectly with the lifecycle of a major data center campus. If you can lock in low-carbon, baseload power for two decades, you insulate yourself from volatile wholesale electricity markets. Luna: And given the political pressure to reduce carbon footprints, nuclear offers a clean alternative to natural gas peaker plants, which are often used to fill the gaps when renewables aren't producing. Lucas: Correct. Natural gas is currently the stopgap solution, but it comes with carbon taxes and price volatility. Nuclear provides stability. The companies facilitating this transition, both in reactor design and fuel supply, are starting to attract serious institutional capital. Luna: It feels like we are witnessing the birth of a new industrial class. Tech, Energy, and Manufacturing are merging. The old silos between sectors are dissolving because the physical constraints apply to everyone equally. Lucas: That is the core thesis. The separation between the digital and the physical world is ending. Every byte of data processed requires joules of energy and cubic feet of space. Ignoring those variables is ignoring the foundation of the house. Luna: So for the listener holding a basket of AI stocks, the question is no longer just about algorithmic superiority, but about operational resilience and resource security. Lucas: Exactly. Resilience is the new moat. If your competitor has better power contracts, they can run their models longer, cheaper, and greener. That is a competitive advantage that code alone cannot replicate. Luna: I wonder if we will see M&A activity accelerate in this space. Larger tech firms buying smaller energy startups to secure supply chains? Lucas: We already are. It is quiet right now because of regulatory scrutiny, but the trend is clear. Vertical integration is the path of least resistance when external supply chains are unreliable. Expect deals to become more frequent as the one trillion dollar spending wave hits full stride. Luna: It is a bold prediction. Integrating energy assets is a massive undertaking. Do you think the market is ready to reward those integrations, or will it punish the complexity? Lucas: The market rewards certainty. Right now, there is uncertainty everywhere regarding energy supply. The companies that provide certainty through ownership or long-term contracts will command a premium. Complexity is only punished if it leads to failure, and these investments are calculated risks. Luna: Calculated risks based on the physical laws of thermodynamics. Harder to game than human psychology. Lucas: Much harder. As we move further into Q4, watch the utility earnings reports closely. Look for guidance on power demand and infrastructure upgrades. That is where the real story is unfolding, away from the flashing screens of the trading floor. Luna: A grounded approach to a head-in-the-clouds industry. I think that is the perfect lens through which to view the rest of this quarter. Lucas: When the next wave of artificial intelligence announcements drops, ask yourself not just what it can do, but what it takes to keep it running. The answer might change how you allocate your capital.