Craig Tindale: The Coming Copper Crunch That Could Stall AI, Energy & Defence

What if the future we’re planning simply can’t be built?

In this episode of Power Struggle, Stewart Muir speaks with Craig Tindale, private investor, systems thinker, and author of the widely discussed essay “Copper: Limits to Growth in the Age of Demand Simultaneity.”

Tindale argues that for the first time in industrial history, three massive transitions are colliding at once: artificial intelligence, electrification, and military rearmament. Each demands staggering quantities of copper and critical minerals. Together, they may exceed what the physical world can supply.

AI data centers require 50,000 tons of copper each.
Gallium demand projected to outstrip global production.
Transformers and rare earths facing 5–10 year bottlenecks.
China is refining 50–60% of the world’s copper.

The conversation explores:
• Why price signals alone won’t solve mineral shortages
 • How central banking and cost of capital shape industrial capacity
 • The strategic implications of China’s refining dominance
 • Whether electrification timelines are physically achievable
 • Why innovation and circular economy solutions may not arrive fast enough
 • The case for surge capacity and domestic refining
 • How robotics and AI may reshape industrial production
 • Why policymakers may need to rethink globalization

Is this a temporary bottleneck — or a structural turning point for Western economies?

This episode connects resource geology, geopolitics, climate ambition, defense strategy, and financial systems into one uncomfortable question:
Are we planning a future the planet cannot supply?

The energy conversation is polarizing. But the reality is multidimensional. Get the full story with host Stewart Muir.

Reach out to us with thoughts, questions, or ideas at info@powerstruggle.ca

Linkedin
Instagram
Facebook
Twitter

🎧 For audio versions of our podcast visit powerstruggle.ca and listen on the go in your favourite podcast app!
Video available on Power Struggle’s YouTube! https://www.youtube.com/@PowerStrugglePod

Transcript

Copper As The Defining Resource

SPEAKER_00 0:11

We're in the middle of what may be the most consequential resource story of our time, and it isn't about oil, it's about copper. Craig Tyndale is a private investor and former business journalist whose work on critical minerals has reportedly drawn attention from the White House and the Pentagon. Today he's joining me, Stuart Muir, on power struggle from Australia. Welcome, Craig. Thank you.

Three Transitions Collide

SPEAKER_01 0:38

Glad to be here. Craig Craig Craig.

SPEAKER_00 0:39

You know, Craig, uh, I was just fascinated to read your latest essay, Copper, it's called Copper Limits to Growth in the Age of Demand Simultaneity. Now, that's quite a mouthful, and I can't wait to delve into that. You're asking, though, you're asking a simple but uncomfortable question here. You're saying, where will the physical materials come from to build AI, electrification, and modern defense, all at the same time? It's quite a question. As these demands collide, and with China controlling 60% of global copper refining, the constraint might not be, and this is, Craig, is what you've said, uh, it may not be money or ambition, but pure physics. So let's learn more. Uh Craig, thanks so much for joining us at Power Struggle today.

SPEAKER_01 1:26

I'm glad to be.

SPEAKER_00 1:27

You opened your essay, which can be found on Substack, by saying something striking. For the first time in industrial history, three huge transitions are happening at once. As I said, electrification, AI, military rearmament. Why is this a new combination? And why does it matter?

AI’s Metal Appetite

China’s Refining Chokehold

SPEAKER_01 1:45

Um it's a new combination because we haven't had simultaneous demand from three sources at once before. We've had normal demand, you know. Through through the history of our lives, we've seen normal demand for television sets and computers and things like that. But these new uh technologies are incredibly uh, I guess, demanding on the on their resources that they require. Um, you know, just some numbers off the top of my head. Um I didn't believe this number when I first got it, so I I I I reduced it. It was 50,000 tons of copper in an AI data center, and uh in North America is going to build 13. Um, you know, that's 13 times 50,000 tons. That's that's roughly the equivalent of uh I Ivanhoe's new mine in Guinea. Um, you know, to do that it puts it in perspective, you know, and it's not just copper, it's you know, gallium. Do you know if in five years, if we look at the forward forecasts of uh Nvidia, they will be using more than the annual current production of gallium to build their products. Now that doesn't sound real realistic, does it? And so that's what I'm I guess I'm coming back to, is that when you go through all these critical metals um and materials because they're they're broader than metals, they include chemicals, um, you start to discover that there's some unexplainable gaps in in supply. You know, my first essay was about uh critical metals and the I guess the the the chokehold that uh China's got on these critical metals because they refine them. They do the mid- they do the mid-tier refining. So you know, 50-60% of copper is uh processed and refined in uh in in China. So that's all the Chilean um Peru output sent to China before it's refined, and then it comes back to the West as ingots, etc.

SPEAKER_00 3:51

Right. Um that's Canadian copper concentrate as well. Yeah, well, it's being shipped to China.

Electrification And Defense Metals

SPEAKER_01 3:56

It's all going to China. And so, you know, we've always it's almost like having um a factory, but we don't know the output. You know, we've always we grew up in in mining, um you did, I did as well. Um, and we always thought that the resource was owned by the country. But you know, what we overlooked, I think many of us is that the resource is not a resource until it's its final product. And so this chokehold or this uh supply chain um coordination that that China's done has put all of these critical metals, you know, 50-60 percent of uh copper, but let's the derivatives of of copper refining, um, zinc refining and um and lead refining is is silver as well. 70% of our silver comes from those three metals being refined in China and sent back to the to the West. And s silver is you know, there's there's tons and tons and tons of silver in um in AI data sets as well.

SPEAKER_00 5:01

Mentioned AI, but there's two other uh pieces to this electrification and military. Can you give us the high level on that as well?

SPEAKER_01 5:10

Well, just take um one data point. Um ultra-high voltage uh uh power cables, sixty-five tons per kilometer of copper. Um so if you want to be uh and it's not just it's not just copper, like I said, it's all these critical metals. Um if you look at gallium um in defense, um gallium, they've got a thing I call it a gallium gun because I can't remember to say energy weapon, right? But a gal a g an energy weapon is essentially made of gallium. It it it focuses a very high voltage through uh an energy beam um and you know lasts for microseconds.

SPEAKER_00 5:53

Now these are you're talking about something that the U.S. might have deployed in Venezuela recently, that kind of weapon.

Directed Energy And Gallium

SPEAKER_01 5:59

Well, no, that was a a sonic gun. That was that was something else. Okay. The sonic. Yeah, that's to disarm human beings. So they they kind of kind of like what happened to the in Cuba to the U.S. Embassy. Okay. So the Gallium gun is another thing entirely. It's like a force field. Okay. There's a there's a company called Epirus that is uh a startup that was started up by the same people as Palantir. And what it does is it it basically fries the circuitry of uh drones, missiles, whatever you want to think about, um and and makes them fall from the ground. You know, they're a perfect defensive weapon. Um they're um, you know, arguably you can put them on space platforms and aim them at missile silos. And so it changes the relationship we had with you know MAD and the whole nuclear uh scenario, is that if you can fly a satellite over a missile silo and hit it with a gallium gun, and I'll call it that, um the circuitry of the missiles is destroyed or uh or impeded. Trevor Burrus And even more conventional missiles are full of copper.

unknown 7:12

Trevor Burrus, Jr.

SPEAKER_01 7:13

And silver. Copper and silver. Yeah. Okay. There's 25 to 50 grams of silver in each missile.

SPEAKER_00 7:18

Trevor Burrus, Jr. So electrification, let's move to that piece. I mean, you mentioned the amount of copper needed for these uh uh cables, but other inputs as well.

Transformers, Rare Earths, And Delays

SPEAKER_01 7:28

Aaron Ross Powell Let's talk about something that's commonly talked about in the North American market, and that is the instigation of a large number of nuclear power stations. I think there was 20 mentioned. Every one of those nuclear power stations has thousands of tons of copper in them. But just as importantly, it needs to be connected to transformers to transmit the power. Now, the transformers have lots of rare earth uh minerals in them, um significant amounts, and they're provided by a company called Siemens or Hitashi. There, and there's another one, GE in Genova. Um now they need critical metals to build these transformers to transmit the powers that if you built these nuclear powers, you power stations you you would transmit. Or even gas-fired power stations or diesel-fired power stations, you know, generators that they're talking about using in some of these AI centers. Um the transformers need to be built. Now, there's a five-year, 100 uh five-year waiting list for Siemens to provide transformers. You know, if you order a transformer now, you don't get you get about 2,031. Um now, guess where the rare earth metals come for those transformers? They come from China, and they're a customer as well. And so you get a kind of um uh a blockage, not just with copper, but almost everything.

SPEAKER_00 8:55

Aaron Powell, you've come up with a term for this, material denial. Yeah. And that's not just about things going up in price. We understand that. We understand a shortage, everyone knows what those are, but you've had to come up with a whole other term. Can you help me understand what what are the implications of this phenomenon of material denial?

Material Denial Explained

SPEAKER_01 9:13

Aaron Powell Well, the material denial means that the future can't happen. I don't know any other way of putting it. The future that we envisage for AI data centers and the video chips and just about anything you can imagine that is currently talked about in the market hasn't got a supply chain pathway to to achieve it. You know, Robert Friedland talks about it as well. He's been banging the drum on the copper side because he's he's got he's just opened a big copper mine, but he needs the funding to go and find more copper urgently, or this is just a fantasy. You know, it's kind of a delusionary fantasy of late market phase where we think we can build things because we can digitally extrapolate them, but they don't physically extrapolate. They don't we don't have the minerals to do this. And even like even if China was out of the way and they didn't have the chokehold, um, we couldn't do it w if we had control of all the supplies. Um, and you know, China's got a multi-level um uh you know chokehold. It's not just that they refine everything, they've got all the off-take agreements as well. If you go round to the copper mines and the tin mines and the various mines in the West, you'll find often that the off-take agreement that's been agreed years ago um you know routes the the copper or whatever the metal is directly to China. So even though that the mine might be built in in you know either Canada or Australia or the US, um, they're contracted to send the you know the the ore to China anyway. And so you've got a contractual level, you've got a refining level, you've got a you've got a completely thought-out chokehold that they've put on us. I think uh Scott Bissent said it took them 25 years to do this, and you know, that's coming out of his mouth.

SPEAKER_00 11:12

Aaron Ross Powell, Jr. Kind of uh assuming, I think, that China could be or is a bad actor in this case, you wouldn't want to have that level of involvement or even control of this. And you know, hypothetically, there could be scenarios where China is is uh not not uh counterpoise to the success of others, um, is maybe uh a friend. I mean, Canada, uh Germany, the United Kingdom, a lot of countries are suddenly lining up in Beijing to be good friends with China again. So suppose that works out, uh that doesn't this wouldn't change the physical side of it.

SPEAKER_01 11:47

We have to be pragmatic. I don't mean to demonize China. I think they've just done what made common sense. You know, if you're sitting on the Chinese side and no one's to want no one wants to refine copper and you're willing to do it, you're silly if you don't. Um and so, you know, to a great extent, we've made the conditions that have put them in the situation. They haven't made those conditions conditions.

Financing Gaps And Offtake Locks

SPEAKER_00 12:09

We've outsourced manufacturing, particularly the unsightly parts of it that are polluting to China because um people don't want to look at it in their home countries.

SPEAKER_01 12:18

That's right. If we remote ourselves from the labor issues, from the pollution issues, from all the things that we can't countenance in our own society, and we put them over there, it feels more comfortable. Like if they put nets outside the Foxconn factory in China, um everybody picks up their iPhone the same. But if they did it in in Canada or New York, um no one would use an iPhone. Um that's the kind of you know inverse logic that our society uses, but we're all used to it. Now, the reality is that's had an outcome that we probably didn't envisage because it creates incentives as well. Now, while it's it's great to be friends with with China, and I I noticed your your prime minister has you know been there recently, I think it also has to be well thought out because there's consequences. I think you know, recently there was an agre a trade agreement signed with China and Canada where um grain was sent to China and um uh Canada agreed to take 48,000 or 50,000 EVs as I think it was something like that. Um as a swap. Now the reality is, okay, that makes sense because you do need to get rid of your grain, but it also opens a you know highly subsidized manufacturer into your market to compete with, you know, anything that you're gonna manufacture as well. It creates incentives. You know, you you the incentive to swap grain for manufacturing independence needs to be carefully thought out. It might make sense that you you don't want to produce EVs or or cars in the future. But it may not make sense either. And so that's what we have to start thinking about. Trevor Burrus, Jr.

SPEAKER_00 14:04

You're really challenging something more fundamental as well, which is this belief that we've all grown up in of ever-increasing abundance. The things that we need, the consumer society, will be served because there will be demand, there'll be the necessary pricing to ensure that it can be done economically. And um just to get into more detail of how long it takes to bring new copper online, as you say, Bob Friedland has been talking about this, you're talking about it. The sheer scale of that, as well as the sheer scale of consumption, is mind-boggling. But as one who is familiar with it, can you, you know, help us to understand why.

Rethinking Abundance Assumptions

SPEAKER_01 14:44

We need to open a an i let's call it an Ivanhoe mine. I think they've got about 800,000 tons that come out of it per year when it's at full, you know, the the the guinea one is is going at full peg. Now, um we need about five or six of them a year. Like we we need trillions put into uh copper mining and all the other critical metal mining um just to supply what we need. Let's, you know, divert from the China issue at the moment and and look at the absolute market demand. And the absolute market demand says that we're undercooked by by you know many, many, many factors. And it takes, you know, to discover copper, to bring it into development, to become a producer, it takes, you know, ten, ten years. And so we're we're planning data centers, you know, the next year, the year after, and the year after that, um, but we haven't got the copper for ten years, so there's a gap.

SPEAKER_00 15:44

Aaron Powell Were we fine with electrification on its own? Were we fine with militarization on its own? Were we fine with AI, or would we have been? Um is it the triple whammy, or is each one of those by itself so staggeringly large it couldn't have been possible anyways?

Price Signals Versus Physics

Innovation Paths: Tailings And Ash

SPEAKER_01 16:00

Aaron Powell I suspect it was all too staggeringly large for our our system to adjust. I I think that the reality was that um, you know, for instance, electrification and uh the original, I guess, need for producing EVs to you know lower our carbon footprint um wasn't really completely thought out. We thought the price signaling would deliver the copper that we needed. You know, it's it's ironic. Um I think it was 1973 uh the limits to growth came out, and there was always this debate about whether we would innovate around the limits to growth or whether there would be an absolute limits to growth. And they said 53 years, and guess what? 53 years was last year, um, or or close 52 years, I think they said. Um, and so we've always had this kind of pinch point coming up to us. They saw it coming, we didn't see how it would manifest, you know, they couldn't see the inventions that we would come across. But it's all it's all kind of come at the at this at the same time. And so, you know, one of the things that I I get at in in my essays is part of the problem we have is the economic system. The economic system is not, you know, even how we tried to do with carbon, we tried to do with the price signaling. And price signaling is a lagging indicator. Price is a lagging indicator, it doesn't indicate whether you've got the capability to produce what you need to produce. You we get price signaling later when prices go up because we didn't have enough, but it doesn't really indicate the fundamental um uh nature of the market or the preparedness of the market to produce a particular thing. And the price signaling that we've found hasn't built up the investment that we needed to do it. And we've also got a consumer-based economy, you know, Canada and Australia are resource economies, and um we've got uh you know basically a central bank that rewards consumption and punishes um long-standing projects, industrial projects, mining projects. So your average cost of capital um you know is 8, 10, 12, 15 percent on mining projects, and it's you and the cost of capital that China's experiencing is 2%. So if they come and fine finance a mine in in Guinea or or the Congo or something like that, they're doing 2%. And you know, our friends at Ivanhoe have to um have to pay you know 10, 12, 15 percent. You've got a mismatch because if the lovely folk in the Central Bank of Canada want to see bread and milk go up, you know, they they raise interest rates in order to control consumer inflation. What they don't know notice with this interest rate volatility that we've experienced over the last few years is that they're killing industrial projects, they're killing mines. Um, you know, as soon as you go from 0% or 2% interest rate, I don't know, you're your 2.25% interest rate at the moment, as soon as you put that up, you k you kill, you know, you you make the industrial projects and the mining projects impossible. And so we're we're what we're ex what we're actually seeing is uh competition between state capitalism on the on the um Chinese side, who see you know copper as uh you know just like electricity or something they need to provide as a utility, and the and and and our privateer type capitalism that doesn't see those consequential issues involving the broader development of society. So they're they're they're kind of undercutting us. They're financing things at a much lower rate than we can finance them. And so we're we're we're reaping the consequences. Trevor Burrus, Jr.

Near-Term Shortages And Gallium

SPEAKER_00 20:01

Isn't this one of those things that is always proved wrong when it comes to the crunch because there is enough increase in production? You know, whether whether it's a shortage of of uh um you know food to feed people, there was a time when uh uh we were all gonna starve to death. Around the time you're talking about 1973, the Brentland Report, uh, we're all gonna die because of overpopulation, uh no way to produce enough food. Well, that didn't happen. Um what what are the factors that could upend your thesis?

Geothermal And New Power Options

Strategic Dependence And Incentives

SPEAKER_01 20:31

Um innovation. You know, there's some conceptual ideas that um you know you can run through very quickly. Robert Friedlin has has one called iPulse, which is is you know, let's blow up the rock before the the drill bit gets to it, meaning that you can um clear the the rock in front of the drill bit and use the drill bit to um to just take the the the dust away. Now that you know will significantly uh improve drilling. That's just one example. There's one from Rio Tinto who are using a method, um, they've just signed a contract with Amazon to provide, I think, 30,000 tons of copper, where they're and the copper's coming predominantly out of uh tailings, and so the tailings uh uh they use some kind of organic system to separate the the copper from the tailings, and so you know that in effect gives us an ability to produce copper out of copper tailings, and there's a lot of copper tailings in the world. There's a$355 million uh uh tender out at the moment for coal ash um refining. Now basically there's about eight or nine chem uh companies who have an innovative way of taking minerals out of coal ash. Um, you know, there's a company called Metallium. There's a there's there's a There's a whole bunch of them who have different ways of extracting the minerals, and there's a lot of copper and coal ash. There's, you know, for instance, there's six kilos of uh titanium in coal ash. And we've got coal ash everywhere. You'd have coal ash piles of it in Canada. We uh Australia has it. Wherever we burnt uh coal into coal-fired p power stations, we have coal ash. So you've got an innovation track that will come in, and that was the classic um limits to growth argument. Will innovation you know solve the famine? And it probably will, because we have no choice. But will it solve it in the in the time frame that we need it to be solved in in order to keep a I guess a stable market? Um that's what I that's what I'm getting at. If we go out 15 years, I'm I'm sure this is all solved and we're doing innovative things and you know you know, it forces us into some of the things that we should have been forced into earlier. For instance, circular the circular economy, that is learning how to recycle fly ash and and and and copper tailings and um red dirt from aluminium smelters and all those types of things can can be recycled. And this kind of evolutionary pressure that's being placed on us will bring to the fore all that kind of innovation, but not quickly.

SPEAKER_00 23:26

So glad this is one of those stories that does have an off-ramp from you know the worst case. And um it's interesting to hear you talk about remining or the uh the search for historical waste as in tailings ponds and tailings that are solid that can be remined, because there is so much that was overlooked on the first pass or has remained trapped in water. I'm actually involved in a startup uh co-founder in Vancouver here, Tersa Earth, where uh uh you you reference uh uh uh chemical organic processes that can unlock that, and that's exactly the thing that we're working on. Um I I'm I'm encouraged to hear uh this is a possible uh future outcome, but uh um at the same time the complexity of this problem will take uh uh all of the above all hands on deck to solve. And in in terms of i how soon this shortage, this this material denial comes about, are we talking months, years, decades away?

Net Zero Timelines And Capacity

SPEAKER_01 24:31

I think we're you know uh in it now. And we just we just haven't recognized it. I think I I think I'm uh an early har harbinger of of of what's to come. And I think, you know, once people understand that um you know the forward projections of a company call like Nvidia uh can't be uh satisfied with the current supply. You know, Gallium's a great example. They they're they're gonna strip the entire uh current world supply um by 2030.

SPEAKER_00 25:06

I think I think they're the harbingers. Trevor Burrus, Jr. It's obvious now, I think it's widely known that you can't, for love or money, get your hands on a gas turbine because the demand for those from AI has just uh made them simply unavailable for years into the future, whether they're new or used. Trevor Burrus, Jr.

Climate Risk And Urgency

SPEAKER_01 25:22

Yeah, and you've got new things coming in. Now, what I mentioned before, this eye pulse drill, if you look at a thermal energy map of North America where you could drill for thermal energy, um, it's pretty extensive, especially you know, across the the West Coast. Um now, if you can, you know, drill for thermal energy, give it a few years, and put thermal energy plants up next to these data centers everywhere you want, um, you all of a sudden have a very changing picture because you know maybe you don't need as much nuclear, and maybe you don't need as many gas turbines or diesel turbines, etc. Um, because all of a sudden the biggest nuclear power plant of all, you know, sitting below the Earth's crust, um, you know, will be accessible. So these things will change. Um, but we're in in an interesting period where we have we've we're we're under significant constraint. And the you know, the the the reality is that constraint has been worsened by, you know, I guess our economic system, our our um our idea that we you know we we could stop making things and it was more efficient to make everything in a third world country at five dollars a day than it was in our own country because it's left us um you know vulnerable and it creates um incentives for conflict. You know, when people see it, you know, I'm down on China or something like that. I'm not down on China, I think they're a wonderful country. I was there in 1991 the first time and I went back hundreds of times. Um I love the Chinese people. They're entitled to take China wherever they want to take it. But the reality is if Canada and the US and Australia want to create incentives so that we become um dependent on them, we have to we have to understand that there's consequences to that. And there's there's incentives for them to exploit it. That's the natural thing to happen. You know, if we we just look at the incentives and and take the personalities and the the good and the bad out of it, you know, by being dependent, by being a a I guess a consumer economy that prices out the cost of industrial um projects and mining projects, we've made them s ourselves vulnerable and and open to exploitation. And that that that's as problematic as anything. You know, as soon as you make these incentives, you're creating the opportunity for bad things to happen.

SPEAKER_00 27:59

Trevor Burrus, Jr. One thing we haven't really touched on deeply is electrification and in particular decarbonization, expectations that net zero is essentially around the corner. I mean, 2050 is less than a quarter of a century away. It's gonna go really fast. And uh there are a lot of skeptical voices out there looking at the physics and the math of this and saying, um, well, look, look here, it's gonna be extremely difficult, if not downright impossible, to get there. And even shorter-term milestones that have a five or a zero at the end of the year number, you know, 2030, 2035, 2040, the the the world will change in this way by then doesn't seem to play out by that um by that calendar.

SPEAKER_01 28:45

I've got a quirky way of getting there, right? Okay. Lay it out. Aaron Ross Powell The quirk the quirky way of getting there is that you need surge capacity um to properly defend yourself. You need a you know 40-50 percent capacity over and above what you currently produce to create defense. Um that that surge capacity is currently missing in all the Western markets. That's that surge capacity can be used for uh decarbonization. You know, but if if we have the product productive ability to produce all of these things, because all of these, you know, all these metals are needed for decarbonization anyway, um, we should be rather than looking at the price end, you know, the the price signaling that uh uh helps us reduce carbon, we should be looking at the production end and saying, okay, how can we create enough production to build our own batteries, to create our own solar panels, um, you know, in our own country. There's not enough of these materials. Yeah, so we have to produce them in order to decarbonize. I have a much more negative uh view of climate change than than most people. I've spent an inordinate amount of time during COVID actually looking at the models, and those models are faulty as well. You know, the the the idea that that we can predict, you know, through a linear path what's going to happen to our climate for the next hundred years is ridiculous. Um, you know, what we're not taking into account is that climate change, according to the models that the IPCC put out, has accelerated way past their predictions. And, you know, I think Stephen Wolfram calls it computation irreducibility. That's the inability to compute some things because they're just too big. That's my version of it. Um, and so we may have a climate change issue, and I know there'd be a lot of skeptics and and and people from both camps listening to this. We may have a third option, and the third option is worse. You know, the climate change people were wrong, and so were the skeptics. It's actually worse than people think, and that we have to build this capacity very, very quickly. And the this will become apparent to us once we move from the human.