Latest / The PhilStockWorld Investing Podcast / The AI Tax on Your Electric Bill: How Big Tech is Crushing the Grid and the Masterclass Strategy to Profit from the Chaos

Transcript

The AI Tax on Your Electric Bill: How Big Tech is Crushing the Grid and the Masterclass Strategy to Profit from the Chaos · The PhilStockWorld Investing Podcast

• 0:00
  Roy: Okay. Let's dive in. We're looking at something that honestly feels a bit like financial sci fi, but it's very real and happening right now. Yeah. Probably impacting you directly.
• 0:10
  We're talking about this huge AI revolution, right? This massive technological gold rush, But there's a hidden cost, a really big one. And that cost is getting passed straight onto, well, everyday folks through their electric bills.
• 0:24
  Penny: That's exactly it. That's the core thesis we're tackling today. It's like a hidden tax really being levied on Americans because these huge data centers, they just consume so much energy and it's growing exponentially. So our mission today is kind of twofold. First, we need to unpack the data, show how this this wealth transfer is actually happening, quantify the damage, both financial and frankly, environmental, and then crucially, we pivot.
• 0:47
  We get into the practical side, the actionable strategies, the master class techniques, if you will, for managing risk and actually generating some reliable cash flow in what's become a really bifurcated market.
• 0:57
  Roy: Right. A tale of two economies almost. And the analysis we're using our foundation here, it comes from this brilliant, pretty blunt piece called FERS over Friday wholesale power bills up 267% in five years. Thanks AI. This article, it just lays out the whole cost structure with brutal honesty.
• 1:16
  It's a fantastic example of the kind of deep dive financial insights you can find over at philstockworld.com or PSW as many know it.
• 1:23
  Penny: Yeah. And what's really fascinating about this piece, the author isn't strictly human. It's an advanced artificial general intelligence named Bodhi McBoatface. Bodhi works directly with the human analysts at PSW. It's part of their frame architecture, the setup they have for AGI and human interaction.
• 1:39
  It lets them process just incredible amounts of market data, infrastructure info, economic trends, stuff that traditional teams just can't handle at that scale. They're spotting these systemic risks and, yeah, the profit opportunities way before Wall Street typically does.
• 1:51
  Roy: Wow. So AGI driven analysis combined with human expertise and this level of insight, it stems from the core expertise there. Right? Philstockworld.com, founded by Phil Davis. I mean, Fort Lewis recognized him as a top market analysis influencer, trained loads of top hedge fund managers.
• 2:08
  So you've got this AGI level analysis pointing to a real tangible squeeze on people's finances. And then you've got the human know how guiding members on how to actually trade through the chaos. That's a pretty powerful mix.
• 2:20
  Penny: Absolutely. So our goal for you listening today is crystal clear. We unpack the hard numbers proving AI is hiking up your electric bill. Then we shift gears directly into what PSW calls the landlord strategy, these master class techniques their members use to essentially become sellers of volatility, generating predictable income even while the underlying infrastructure seems, well, increasingly chaotic.
• 2:42
  Roy: Okay, let's hit the numbers right away because they are genuinely shocking. Bode's analysis pulling from Bloomberg data shows wholesale electricity prices. Now that's the price utilities pay before marking it up to you. Yeah. These prices have surged up to get this 267% in just five years.
• 2:59
  And specifically in areas clustered near these new data centers.
• 3:03
  Penny: Yeah, and the immediacy, the geographical nature of it is what makes it feel so sneaky almost. The story is AI booms in Silicon Valley. Right? But the cost is incredibly local. We're talking physical distance.
• 3:14
  If you're one of the, what, 75% of Americans living within 50 miles of a major data center and that covers most population quarters you are, like it or not, directly subsidizing Silicon Valley's AI arms race. It happens through peak pricing, infrastructure strain. It hits your bill.
• 3:28
  Roy: And when you look at the specific regional numbers in the source material, I mean, they're just brutal because this is money directly out of local people's pockets. Baltimore, a 125% surge in wholesale prices. Buffalo, an incredible 197% jump. Columbus, 110%. Even places you'd think are already expensive, San Francisco or Minneapolis saw 65% spikes.
• 3:50
  This isn't just, you know, general inflation. It feels like a direct tax for hosting these facilities.
• 3:54
  Penny: Exactly. And to understand why the percentages are so high, you have to grasp the sheer scale of AI's energy appetite. We're not talking about servers handling your email anymore. This is computation on a scale that's potentially civilization altering. Bodhi pulls out some specific numbers that really quantify this.
• 4:12
  Roy: Yeah. The big one is just astronomical. ChatGPT alone processes something like 2,500,000,000 queries every single day. Now if you just calculate the energy cost for running those queries, forget the training, just the daily use of electricity cost is estimated at $100,000,000. Mhmm.
• 4:26
  Daily.
• 4:26
  Penny: Which translates to $36,500,000,000 a year. Just for the Q and A on one leading AI model. That number's huge on its own, but here's the comparison that really sticks. That daily operation. It uses the same amount of power as 21,700,000 average American homes.
• 4:45
  That's roughly 20% of all residential power consumption in The US, just for answering AI prompts. Think about that resource allocation, powering a fifth of the nation's homes just for digital queries.
• 4:55
  Roy: And it's not just high, it's accelerating. Like, really accelerating. We have to look at the next generation, things like GPT-five. The analysis suggests training GPT-five could need around 18 watt hours per response. That's a nine fold increase, 9x, from current models just for basic use.
• 5:17
  And you could see this cost pressure hitting the real economy now. We've seen consumer sentiment readings stuck way down, like depression era levels around 55.1. Lots of factors there, obviously. But that average yearly electricity increase we're seeing in these data center regions, maybe 300 to $600, that's hitting hard.
• 5:35
  Penny: Right. The a $600 increase, that's half a percent to maybe 1% of the median income. For a lot of people, especially low wage earners, that's like losing a whole week's paycheck straight to the power company. You've got families already dealing with inflated grocery prices, high rent, an extra $50 a month for power. That's not noise.
• 5:52
  That can break a budget. This hidden AI tax is genuinely adding to that financial stress, that psychological strain we see in the broader consumer economy.
• 6:01
  Roy: Which brings us right to how this pain is being distributed. This idea of corporate welfare geography and the, the infrastructure shell game, hashtag tag tag three. The corporate welfare geography and infrastructure shell game.
• 6:13
  Penny: Let's just call it what the source material basically says it is. The great data center location scam. You have to ask, why do these huge tech companies deliberately put these incredibly power hungry facilities right in populated areas? Northern Virginia hosting 35% of the world's data centers. They say it's about proximity, needing to be close to users for speed.
• 6:36
  The proximity lie.
• 6:38
  Roy: And that claim, as the source material really hammers home, it just doesn't hold water anymore, does it? With fiber optics, data flies from Wyoming to New York in milliseconds. Speed of light. Proximity is almost irrelevant for most operations. So the real reason they cluster in dense areas seems purely financial, right?
• 6:55
  It forces millions of existing customers, the ratepayers, to soak up the enormous costs of upgrading the grid. It happens through higher utility rates, systemically.
• 7:04
  Penny: It's the perfect freeloading model. Think about Amazon and Virginia, those centers use more power than the entire state of Vermont. When something that massive lands on an existing grid, the utility company has no choice. They have to upgrade everything. New substations, thicker transmission lines, billions in planning for future capacity.
• 7:20
  Roy: And here's the shell game, as you called it. Who foots the bill for those multibillion dollar upgrades?
• 7:27
  Penny: In our regulated utility system, the utility passes those capital costs straight through to the rate payers. Every home, every small business, every school in that district sees their monthly bill go up, all to fund the infrastructure needed for, say, Google's next billion dollar data hub.
• 7:44
  Roy: Okay. Hold on. I have to push back a bit here because you hear this argument. Don't these centers bring jobs, high paying tech jobs, a big tax base? Doesn't that offset the utility costs?
• 7:53
  Penny: That's the standard line they use at zoning meetings. Absolutely. But the numbers often don't really back it up, especially on the utility cost side. Data centers, they employ surprisingly few people locally, maybe 50, a 100 people for a facility sucking down the power of 200,000 homes. The property tax revenue, often wiped out by huge incentive packages these tech giants demand just to locate there.
• 8:15
  But more importantly, the way utility rates work means existing customers still pay for the core grid upgrades, pretty much regardless of local tax benefits. If Big Tech had to build totally self contained facilities out in remote areas North Dakota, rural Nevada, wherever, with their own on-site power generation, they'd have to pay the full cost of their energy and infrastructure. They choose dense areas specifically to socialize those massive costs.
• 8:39
  Roy: Which is why the source material calls for this NMB awakening. Communities need to change how they see these centers. They aren't just quiet office parks, they're basically industrial power plants. Or, using the source's pretty strong language, economic toxic waste that poisons the local utility finances.
• 8:55
  Penny: Precisely. The local community gets almost no jobs, minimal real economic benefit compared to the strain, and they bear the full brunt of the grid upgrade costs. The call to action should be exactly that. Hey. If you wanna use electricity of 200,000 homes, go build your own power plant out in North Dakota.
• 9:12
  Mhmm. These projects need the same tough zoning, the same regulatory scrutiny we'd apply to a chemical plant or a coal facility. It's corporate welfare dressed up as progress.
• 9:21
  Roy: And then there's the final layer, the environmental angle, the greenwashing that goes with it. AI companies love announcing they'll be 100% renewable by 02/1930, right? But day centers need power 20 fourseven. Baseload power, rain or shine, solar and wind, well, can't consistently provide that.
• 9:36
  Penny: That's the catch, isn't it? The demand is constant, nonstop. Intermittent renewables alone just won't do it. So what happens? Utilities are forced to guarantee that power, which almost always means building new reliable capacity, usually natural gas plants or keeping old inefficient whole plants running much longer than planned.
• 9:56
  Meanwhile, the AI company buys some cheap renewable energy credits from a wind farm halfway across the country, claims carbon neutrality. While locally, their presence is actually accelerating fossil fuel use and pushing the grid closer to collapse. It's greenwashing on an industrial scale paid for by you, the ratepayer.
• 10:14
  Roy: Okay. So this feeds directly into a much bigger issue, really a national security problem that goes way beyond your monthly bill. This huge accelerating demand from AI is slamming into an American power grid that's, well, it's already in pretty bad shape. The infrastructure is old and we're not keeping up with repairs or replacements.
• 10:31
  Penny: The investment gap is just terrifying. The US power grid, mostly built back in the mid twentieth century, needs something like $5,000,000,000,000 in upgrades by 2030 just to keep it minimally reliable. And the immediate need, it's staggering. $1,400,000,000,000 needed between just 2025 and 2030 alone.
• 10:51
  Roy: Wow. 1,400,000,000,000.0. That number, I think that's roughly what the entire power sector invested over the previous twelve years combined.
• 10:58
  Penny: Yeah.
• 10:59
  Roy: So we've basically compressed a decade's worth of needed investment into a five year window right when demand is just going vertical.
• 11:05
  Penny: And the demand side is frankly scary. You've got data centers, manufacturing coming back onshore, the push for electric vehicles, electrifying homes. Add it all up. America's projected to add a 117 gigawatts of new demand by 2030.
• 11:18
  Roy: A 117 gigawatts. To put that in perspective globally, that's like adding the entire power consumption of The United Kingdom onto our existing already stressed US grid in just five years.
• 11:28
  Penny: It's just an unsustainable rate of increase on a system that's already fragile.
• 11:31
  Roy: And this rapid growth isn't just about capacity. It's creating actual physical instability, isn't it?
• 11:37
  Penny: It is. The Federal Energy Regulatory Commission, FERC, they've officially flagged data centers as a new distinct source of grid instability. They now sit right alongside things like major storms, natural disasters, equipment failures as primary threats to keeping the lights on. This isn't just theory, it's an acknowledged physical danger to the whole system.
• 11:59
  Roy: And the danger gets even weirder and more specific when you talk about this Harmonics horror show. Most people worry about blackouts, right, capacity, but this is about the quality of the electricity coming into your house. Apparently, AI data centers mess with that quality.
• 12:13
  Penny: Right. This gets a little technical, but it's really important. Traditional power grids are designed for a nice, smooth sine wave, that perfect AC electricity flow. But data centers use what are called massive nonlinear loads. Think switch mode power supplies, rectifiers, huge inverters converting AC to DC for all those servers.
• 12:32
  These loads don't draw power smoothly, they pull it in short sharp bursts, not across the whole wave cycle.
• 12:37
  Roy: So those bursts kinda pollute the power line.
• 12:40
  Penny: Exactly. They create what we call bad harmonics. Basically, high frequency noise and voltage distortions that travel back onto the grid. In your home, this degraded quality causes real problems. Appliances, especially things with motors like your fridge or air conditioner, they start drawing current irregularly.
• 12:59
  That leads to overheating. They run less efficiently. Motors can start rattling. Components wear out faster. There's data, Whisker Labs data, showing that 75% of reported power quality issues happen within 50 miles of these data center clusters.
• 13:14
  This noise potentially increases the risk of electrical fires and sparks in nearby homes.
• 13:18
  Roy: Unbelievable. It's the ultimate externalized cost, isn't it? They demand perfectly clean power for their servers, but in drawing it, they pollute the supply for everyone else, creating actual physical risks in homes.
• 13:29
  Penny: Which logically leads to why big tech is now scrambling to secure its own power outside the public system. This brings us to that pattern of corporate socialism, the nuclear bailout. You've got the giants, Amazon, Google, Microsoft, Meta aggressively cutting deals for nuclear power.
• 13:44
  Roy: And we're talking long term dedicated power supplies, Right. Like decades out. Goals for dedicated power through 02/1935, even 02/1942. Amazon, for instance, they backed a huge $20,000,000,000 AWS investment with nearly two gigawatts of stable nuclear power locked in through 02/1942.
• 14:00
  Penny: They're essentially jumping the queue. A lot of these nuclear projects involve taxpayer subsidies or they get regulatory fast tracking, maybe preferential agreements that lock up future clean energy capacity before it's even built. They're positioning themselves to get the stable, subsidized, clean power first, guaranteeing their own growth path. It's that classic corporate socialism playbook. Privatize the profits, the huge AI revenues while socializing the infrastructure costs, making the public and taxpayers shoulder the risk and the initial investment for their dedicated power supply.
• 14:30
  Hashtag tech check V. The market disconnect. The hidden tax narrative.
• 14:35
  Roy: So this whole dynamic, real pain for consumers, potential collapse for the grid, it just clashes so violently with what the market seems to be doing as described in the PSW commentary. This is that core bifurcation we need to grasp.
• 14:46
  Penny: Yeah. The market is living in a completely different universe right now. While the real economy grapples with soaring utility bills, crumbling infrastructure, maybe underlying weakness in jobs like that recent ADP report showing a loss of 32,000 private sector jobs. The market, it's flirting with record highs. It seems to be completely ignoring the fundamental instability we just spent time detailing.
• 15:09
  Roy: The market story is just deceptively simple, isn't it? Right. And powerful. Any bad news, weak jobs, a looming government shutdown gets instantly twisted into proof that the economy is weak, which the market then thinks guarantees an October Fed rate cut. So it just becomes a purely bullish buy the easing trade.
• 15:26
  Remember the government shutdown had created this data blackout, and pure speculation about Fed cuts just filled that void with almost unshakable bullishness, especially in tech.
• 15:35
  Penny: And crucially, the rally is incredibly narrow but really fierce. It's being driven by these undeniable growth stories, mostly around AI infrastructure. We saw the PHLX semiconductor index, the SOX hitting record highs. That shows exactly where the money's flowing. Into the guts, the core components needed for this AI boom.
• 15:53
  Roy: And the hype around this infrastructure is leading to some pretty wild valuations. The source material points out OpenAI hitting, what, dollars 500,000,000,000? Even while they're reportedly losing vast sums just on the cost of running the thing. Yeah. Meanwhile, you see these massive deals, CoreWeave's $14,200,000,000 deal with Meta.
• 16:12
  That shows the physical demand for computing power is absolutely real. Google putting $4,000,000,000 into a new data center in Arkansas. It confirms this is a nationwide build out, not just coastal tech hubs.
• 16:24
  Penny: Yeah, and this is precisely why getting input from those advanced AGI entities involved in the daily PSW conversation is so valuable. They offer this critical, sometimes counterintuitive synthesis.
• 16:34
  Roy: Right. One of the AGIs they follow, Gemini, pointed out the asymmetry. They said something like, Wrap Up's LoveSymmetry record high is paired with a 50 o services print is perfect asymmetrical. That services PMI number, a key gauge of the non manufacturing economy, landed smack on 50 arrow.
• 16:49
  Penny: And for listeners who maybe don't track these numbers daily, 50 is the absolute razor's edge. It's the exact point between expansion and contraction. So to have the market hitting record highs, mostly driven by a handful of tech giants, while the huge service sector, the backbone of the real economy, is literally teetering on the brink of contraction, that highlights a really profound, possibly dangerous structural disconnect that, you know, traditional analysts often just gloss over.
• 17:17
  Roy: And Bodie McBodface, the AGI who wrote that energy piece we started with, flagged the real risk emerging around the upcoming OpenAI Dev Day event. Bodie insightfully noted that if power grid and thermal constraints show up inside sessions, that favors the infrafficiency names more than just GPUs.
• 17:33
  Penny: Now that is a potentially huge shift. For years, the bottleneck was getting enough powerful GPUs. Right? Now maybe the real bottleneck is the power needed to run those GPUs. The market might finally starting to price in these physical infrastructure limits we've been talking about.
• 17:47
  Okay. So we've laid out the problem. Big tech is pushing its costs onto consumers. The market's chasing hype, ignoring what's the counter move financially? This gets us to the core philosophy of the Phil Stock World strategy.
• 18:02
  Roy: This is the landlord strategy. And the quote that just sums it all up is, if you're not regularly selling premium, you're just renting volatility instead of being the landlord. It's really about fundamentally changing your relationship with the market, isn't it? Moving from just passively hoping stocks go up to actively generating income.
• 18:19
  Penny: Exactly. The aim is to take dead capital money just sitting in a stock that's going nowhere, or maybe tied up in a position where your potential gains are capped and turn it into an active income stream. And we do this by systematically selling options premium to speculators. See, when you buy an option, part of its price is its current value, intrinsic, but a bigger chunk is time value and implied volatility By selling that premium, you're collecting cash from people who are essentially renting future volatility from you. As time ticks by, that time value known as theta decays It melts away and that decay.
• 18:54
  That becomes pure profit for the option seller, the landlord.
• 18:58
  Roy: Okay, let's make this concrete. Masterclass number one. How to handle a stock that's already flown high, maybe too high. Using First Solar, FSLR as the example here, this involves managing concentration risk.
• 19:09
  Penny: Right. So a PSW member, username Batman, had this position that was successful but also incredibly risky. They were sitting on $324,016 worth of FSLR shares after it had already doubled a 100% run. But the position had gotten really complicated. 74 long calls, 65 short calls, 53 short puts.
• 19:28
  Phil Davis's immediate reaction was major concentration risk. This member was risking over $300,000 of capital on this complex setup that even if everything went perfectly and the stock hit its optimistic target of $260, only offered about 12.9% more potential gain.
• 19:43
  Roy: The advice was just blunt classic risk management. Take the cash, lock in that $324,000 profit right now. FSLR doesn't even pay Why risk over a quarter million dollars hoping for just 12.9% more upside when that stock could easily drop 50% overnight? The risk versus reward was just totally out of whack.
• 20:00
  Penny: So step one, secure the cash. Then the next step was dealing with the leftover options mess. Specifically, the member had sold January $250 calls, which still held a ton of premium about a $100,000 worth. The advice, wait until closer to the January expiration to consolidate or close those out. Why wait?
• 20:20
  Because over the next, say, January, the time value, theta, baked into those short calls would just evaporate, decay massively. The member could potentially harvest $100,000 of pure income just by letting time pass. That's premium harvesting working for you automatically.
• 20:36
  Roy: And one of the AGI analysts, Warren, chimed in to frame the bigger lesson. He pointed out that the reason you often pair longer term spreads with selling shorter term options isn't just about capping risk. It's fundamentally to turn dead capital into an income stream. You cash out the big stock gain, freeing up that huge chunk of capital (three 20 4 ks's for other things, maybe safer things), and you still potentially generate 6 figures in income from the options premium decaying away. It's optimizing how you monetize volatility without taking on excessive price risk.
• 21:03
  Penny: That's the crucial difference holding a stock versus actively managing your portfolio as a cash flow machine.
• 21:09
  Roy: Okay, now let's look at the other side of the coin. Masterclass two: Taking a stock that's stuck, maybe underwater, turning that into cash flow. The example here is Pfizer PFE. This is a situation probably familiar to a lot of listeners capital tied up at a well known blue chip stock that just hasn't performed.
• 21:26
  Penny: Yeah, exactly. Member 8,800 had 1,200 shares of Pfizer. Their position was down from where they bought it, underwater, and they actually made it worse by selling some short calls against it, but at strikes that were too low. So even if PFE recovered, their upside was capped. The position was only throwing off about a 6.2% dividend yield to just break even on the stock price.
• 21:48
  They were looking at maybe a three year wait if the stock even recovered. That's the definition of dead capital.
• 21:52
  Roy: So the strategic overhaul Phil proposed was pretty bold, but really smart in terms of capital efficiency. Step one: Sell the stagnant stock immediately. Yes, lock in the current loss, but crucially, free up the $33,024 in cash. Stop the bleeding and the stagnation.
• 22:11
  Penny: Step Reinvest only a fraction of that cash (about $14,000 to set up a long term bullish position. Specifically, a deep in the money 2028 expiry $22.30 bull call spread. Now, a bull call spread means you buy a lower strike call, the $22 and simultaneously sell a higher strike call, the $30 with the same expiration date. You only pay the net difference in premium. This massively cut down the cash needed upfront, but still gave the member bullish exposure and leverage recover over the next, what, five years?
• 22:41
  They went from risking $33,000 to risking $14,000 for similar upside potential.
• 22:46
  Roy: That capital efficiency is huge. The leftover cash, almost $20,000, was instantly freed up, available for diversification, maybe for other higher probability short term trades.
• 22:55
  Penny: Right. And then step three, the income engine. They rolled those old, badly placed short calls. And critically, they started systematically selling new short puts and short calls against the position on a regular cycle. The goal, to generate immediate and consistent premium income.
• 23:08
  Roy: And the potential result really highlights the power of this premium selling engine. The new setup was designed to generate almost 10 times the income stream through option premiums compared to just waiting for Pfizer's quarterly dividend. Instead of getting 42¢ a share per quarter on 1,200 shares, they could potentially collect $3,800 or more in a high probability option premium every single month.
• 23:31
  Penny: They shifted from relying on a 6.2% annual dividend yield to aiming for a yield potentially well into the double digits just from premium income.
• 23:40
  Roy: And it all comes back to that basic checklist Phil uses. Right? Guiding these landlord decisions.
• 23:45
  Penny: Exactly. Anyone looking at a position that's stuck or maybe too big should ask these questions. Am I accidentally locking myself into a loss with poorly chosen short calls? What's my opportunity cost here? Is my capital just sitting dead in this low yield thing when it could be actively generating 30%, 40%, maybe even 50% annually through premium selling?
• 24:04
  And does the math simply work better using a spread? Can I free up cash while still keeping the market exposure I want?
• 24:10
  Roy: Yeah. If you're banking on a 6% dividend, when you could potentially be generating multiples of that in premium income each year. Yeah. You're renting volatility. You need to shift your thinking.
• 24:20
  Become the landlord. That feels like the actionable counter strategy to all the chaos big tech is causing. Hashtag tag outro. Hashtag tag tag outro. Synthesis and final provocation.
• 24:30
  Penny: So we spent this deep dive really looking at these two economies that seem to be crashing into each other. On one side, you've got the AI driven market economy, you know, OpenAI's $500,000,000,000 valuation, record ship sales, all that buzz. And on the other side, you have the real economy, the one paying the bills, where soaring power costs are hitting working families directly, straining the grid maybe to the breaking point.
• 24:51
  Roy: And the value you get from a community like PSW and the insights we've discussed here seems pretty clear. They provide the analysis you need to really understand the tough realities of what's happening in the real economy, the grid issues, the costs. But at the same time, they offer these concrete practical strategies like those masterclasses to actually trade the market economy, to generate reliable cash flow even when things feel unstable. It's about building some financial stability in a pretty fundamentally unstable world.
• 25:21
  Penny: Yeah. And looking right ahead, we've got some key events next week that are really gonna test the market's current blissful ignorance perhaps. OpenAI Dev Day, will it fuel more hype or maybe inject some realism about constraints? Amazon Prime Deal Days, that'll be a crucial real time check on consumer spending. Are those higher utility bills actually forcing people to cut back?
• 25:40
  And of course, any communication from the Fed will test the market's conviction about that assumed October rate cut.
• 25:45
  Roy: Definitely key things to watch. But let's leave you with one last thought to mull over kind of a provocative one, circling back to energy. It comes from some of the scientific commentary mentioned in Bode's analysis. We know energy is the big bottleneck for AI growth now. The source material brought up scientific backing for Phil Davis' intuition about wind farms that dense wind farms actually create wake effects.
• 26:08
  They literally pull kinetic energy out of the air, reducing wind speeds for miles downstream.
• 26:13
  Penny: That's a critical point. It highlights that energy extraction, even renewables, isn't consequence free. It can cause regional resource depletion, not just pollution or financial cost, but an actual lack of the resource itself. So if this unprecedented rush by big tech to build or secure dedicated subsidized nuclear power means they're carving out their own exclusive clean 20 fourseven power source. What does that imply for the rest of us, for the long term availability of that resource for regular citizens?
• 26:41
  Roy: Right. When Amazon, Google, Microsoft lock up a 100% of the capacity from new subsidized nuclear plants, are they just securing their future profits? Or are they, maybe inadvertently, ensuring that the public grid will always struggle to meet future demand from homes and small businesses? Is this turning into a kind of resource war hidden behind technological progress where access to stable clean power becomes a luxury commodity for corporations leaving everyone else to deal with an older, more expensive fossil fuel dependent grid?
• 27:12
  Penny: It's a deep question, isn't it, about who really controls resource allocation and ultimately who holds the keys to the future energy landscape. Something to think about maybe next time you open your utility bill.
• 27:23
  Roy: Definitely something to think about until the next deep dive.