AIと経済の変革
Welcome to the Depth Dive. It's early morning, July 2nd, 2025, just before 6am here in Japan.
And the markets are definitely moving. The Nikkei, you know, Japan's main stock index, it's down quite a bit in night trading, sitting around 39,575 yen.
And all this is happening with, well, lots of noise around Trump, Elon Musk, and tariffs. Always tariffs.
Now, our latest look at the sources suggests these Trump tariffs, the import duties, they're probably sticking around.
Even after that 90-day suspension thing, likely staying at those April 2nd levels.
So for this Deep Dive, we're asking, you know, what's the real goal here? What are these tariffs actually trying to achieve?
And that 90-day window, was that ever truly about negotiation? Or maybe something else entirely?
Yeah, those are the big questions, aren't they? And today we're going to try and connect some dots.
Dots that might seem pretty far apart at first glance. We're talking current economic shifts, the explosion of AI, and even some classic economic ideas.
History, too. The aim is to pull back the curtain on these tariffs, give you perspective that goes way beyond just the headlines. It's a different way of looking at it.
OK, let's unpack that. The sources really point to 2022 as a key moment for AI. Its arrival wasn't just gradual, it was transformative.
And companies now are really wrestling with this. How do you take this incredible AI growth and actually turn it into revenue, make it part of the business model?
Exactly. This AI transformation, it's not just jargon. The sources frame it as a fundamental driver for the global economy, almost like a new industrial revolution, you could say.
And there's a clear pattern emerging, using AI to replace human labor. It's about cutting personnel costs, boosting efficiency. That's a major trend.
AIの電力需要の増大
So if AI is this massive engine changing everything, what does it absolutely need to run? What's the fuel? It's like lifeblood.
That's the absolute crux of it. And the sources are crystal clear. Electricity, power is becoming the literal bloodstream for AI. Think about it, humans need blood, right? AI needs power. Massive amounts of it, constantly.
And managing this huge growing appetite for electricity. That's not just a technical issue, it's shaping up to be a critical global economic challenge.
Which leads to a very practical problem, a big one. Building new power plants, upgrading the grid. That stuff takes time. The sources say, what, two, three years?
Minimum. So you've got this instant, massive demand from AI.
Right, insatiable demand now.
But the supply of new energy is on this much slower track.
Precisely. Which forces the question, if we can't just instantly build more power stations, how do we find the extra electricity we need right now? For AI's immediate growth spurt, this isn't a future problem, it's here, today.
OK, this challenge, it brings up an interesting pattern the sources highlight, a historical one. It seems like big economic breakthroughs, major innovations, they often don't just pop up when everything's booming, sometimes they come out of slowdowns, or even periods of, well, disruption, destruction, some call it.
Like, think about the Internet. It really exploded after the dot-com bust, right? Or social media giants, Facebook, Amazon's growth, Twitter, a lot of that ramped up after the Lehman shock in 2008, that massive financial crisis.
Exactly. And what's really interesting is how this echoes economic theory. Specifically, Joseph Schumpeter's idea, you might have heard of it, creative destruction. Schumpeter argued that real innovation often needs the old ways, the existing structures, to be disrupted, broken down even, to make space for the new, better stuff. So applying that thinking, maybe AI's big leap forward also needs some kind of disruption, a shake-up to really take hold.
Now, let's bring this back to that electricity problem. Remember that data point. During the Lehman crisis, electricity use dropped globally by about 5% on average over roughly 17 months. So here's the hypothesis from the sources, and it's pretty bold. What if you could suppress electricity demand by just 5% for maybe two years? The idea is our current power grid could actually handle the extra load from AI. We could generate that needed surplus without building anything new right away.
OK. A 5% drop in electricity use freeing up capacity for AI. But how do Trump tariffs fit in? That seems like a huge leap. Trade policy impacting energy grids. How does that work?
Right. This is where it gets really interesting, where the pieces connect in an unexpected way. The stated goal of these tariffs, OK, slow down global trade. That's what they're designed to do. They act like a wall, a breakwater, slowing down goods flowing into the U.S. And the U.S. is, you know, the world's biggest consumer market. So this immediately adds costs all through the supply chain.
Manufacturers then, they get nervous. They don't just produce to fill factories anymore. They only make what they're pretty sure they can sell. And consumers, well, stuff gets more expensive because of the tariffs. So they tighten their belts by only what they really need, not the extras.
OK. So you've got this double break effect, less manufacturing, less buying, a general slowdown. Got it. But the electricity link, how does that directly free up power?
That's the subtle but critical connection the sources are pointing to. This general economic slowdown, it directly pushes down electricity use in existing factories. Fewer production lines running, maybe. Machines sitting idle more often, maybe even cutting back shifts. All those things save power and that saves electricity. That is the surplus. It's electricity that was being used for traditional industry now freed up, ready to be redirected to AI, to data centers.
AIの成長を支える戦略
Wow. OK. That is quite the hypothesis. So the argument is Trump's real goal, maybe the hidden goal with the tariffs, isn't just trade balance. It's to deliberately engineer a slowdown to create the conditions for AI to take off immediately, like a controlled demolition to clear ground for new construction.
That's the interpretation presented. It's about rerouting that bloodstream, the electricity, to AI. Because without that surplus, if AI tries to grow too fast on the current grid, you could get power shortages, real ones. And that could trigger massive energy inflation, choking off the very innovation you want. So slowing down the existing economy with tariffs, it's framed as a strategic play, a deliberate move to generate the needed power surplus. So AI and data centers can hit the ground running.
And going back to Schumpeter, that fits perfectly, doesn't it? Creative destruction, slowing down, maybe even damaging the old industrial energy usage to fuel the new tech revolution.
It really does align with that idea. And it also makes you rethink that 90-day negotiation period. Maybe it wasn't really about haggling over percentages on steel or whatever. Perhaps it was more about internal discussions, aligning behind the scenes, letting this deeper strategic goal settle in among those involved, away from the public eye.
The public reason might not have been the full picture.
What a fascinating way to look at it. This whole deep dive suggests something pretty counterintuitive, that AI's bright future, this huge innovation wave, it might actually depend on policies that intentionally put the brakes on the economy we have now, just to free up the power it needs.
Absolutely. It challenges the surface narrative, doesn't it? What looks like just another trade spat could, under this interpretation, be a calculated maneuver, a way to accelerate the next big technological leap.
And it leaves you wondering, doesn't it? How might you, listening to this, start looking for similar hidden gears, these kinds of strategies in other global events, connecting those seemingly unrelated dots? Sometimes the real story, the deeper purpose, isn't the one right in front of us. The world is often more layered than we think.
