원문 보기
The AI to Bitcoin rotation is here. And I don't mean it's coming. I mean it's already happening quietly underneath the price chart while most people are still staring at the wrong number. For months, Bitcoin's price has been lagging behind. AI captured all the money. AI captured the headlines. AI captured every ounce of momentum in the market and Bitcoin Bitcoin just sat there watching from the sidelines.
But over the last 6 weeks, that relationship has started to flip. And this matters to you no matter which side of the trade you're on. If you own Bitcoin, you need to know whether the lag is finally ending. If you own AI stocks, you need to know if capital is actually leaving those positions, and if it is, where it's going next.
And if you're sitting in cash, waiting for one more confirmation before you make a move. You need to understand the price of being late, because rotations like this one don't just move markets. They decide which assets compound your wealth for the next decade, and which ones quietly get left behind.
Here's where it starts. Over the last 60 days, the largest Bitcoin wallets in the world added close to $3 billion worth of Bitcoin. That's a big number on its own. But then we got the signal that made everyone sit up. I'm talking about Stanley Ducken Miller. His family office just made the loudest move we've seen all year.
They disclosed $125 million position. The GOAT moved. And normally that's all the information you'd need. Except he didn't move where everyone assumed he would. Smart money almost never makes the obvious trade and Draen Miller is proof of that. He's moving through what I call a hidden bridge. A bridge that connects the AI boom and the Bitcoin boom at the exact same time.
So, in this video, I'm going to break the whole thing down. I'll show you the hidden bridge Draen Miller is walking across, explain why it matters for both Bitcoin and AI, and then walk you through how to confirm this rotation for yourself before the rest of the crowd figures it out. Let's get into it.
If you don't already know who Stanley Ducken Miller is, here's the short version. For more than 30 years, running some of the largest funds on the planet, he never once had a losing year. Over three decades, he averaged returns above 30% annually. He was the lead portfolio manager for George Soros's Quantum Fund, and he was at the center of one of the most legendary trades in financial history, the 1992 short against the British pound that broke the Bank of England and made the fund over a billion dollars in a single day.
That track record is why people call him the greatest to ever do it. But the reason his most recent filing matters isn't just the resume. It's the way he operates. Draen Miller has always been open about how he thinks. He doesn't fall in love with a position. He doesn't become an idologue about any single trade.
When the facts change, he changes with them immediately without ego. That's the part of him that makes this filing worth paying attention to. On August 14th, his family office disclosed its 13F, the quarterly filing that shows what a fund held at the end of the period, in this case as of June 30th. And what it revealed was four positions in Bitcoin mining companies.
Bit Deer Hut 8, Riot Platforms, and Iris Energy. All four were plain stock positions worth roughly $125 million combined. At the same time, a handful of long-held names disappeared from the portfolio entirely. Broadcom, Intel, and Micron were gone. Now, here's where most people online got it wrong. Second, this filing dropped.
You had commentators on YouTube and on X-Racing to the same lazy headline. Draen Miller dumped AI and rotated into Bitcoin. That's not what actually happened. Look past the headline and the picture is more nuanced. He didn't abandon AI at all. Taiwan Semiconductor remained one of his largest holdings in the same filing.
And in that very filing, he added new positions in AMD, Lamb Research, Equinex, and Alphabet. This wasn't an exit from AI. It was a rotation within the broader theme, a repositioning towards something underneath both booms at once. He skipped the obvious Bitcoin trade entirely. He didn't buy spot Bitcoin and he didn't load up on a Bitcoin ETF.
Instead, he picked four companies sitting precisely between the AI economy and the Bitcoin economy. Companies that control a resource both industries are fighting over right now. That's the detail that makes the entire thesis click into place. But before we go further into the bridge itself, you have to stop watching the obvious number.
And the obvious number is Bitcoin's price. Every single day, people pull up the chart and ask the same question. Is it moving yet? Did it break out? Has the rotation started? Here's the problem with that approach. Price is almost always the last place a rotation becomes visible. By the time price confirms the move, the early money is already positioned, and everyone watching the chart is left paying what I call the confirmation price, a much higher entry for much lower conviction.
If you actually want to see a rotation before the crowd does, you have to look underneath the price, not at it. So, let's start with onchain data. Through August 9th, over roughly 60 days, CryptoQuan's cohort of wallets holding more than 10,000 Bitcoin, added about 46,420 Bitcoin, worth close to $2.9 billion at current prices.
At the exact same time, the smallest wallets, the ones holding between 0.1 and one Bitcoin, reduced their holdings by roughly 9,700 Bitcoin. Read that carefully. Retail is selling directly into the hands of Wales. That's not a healthy pattern to be on the wrong side of, and it was happening while everyone assumed Bitcoin was dead money, destined for one more leg down.
While retail sold, Wales accumulated quietly before the price gave anyone the confirmation they were waiting for. It's not only visible on chain either. Look at the institutional filings. Jane Street, one of the largest market makers on the planet, disclosed a $291 million long spot Bitcoin ETF position at the end of the first quarter.
By the end of the second quarter, that position had grown to nearly $940 million, more than triple in a single quarter. Jane Street isn't placing a casual directional bet. They're building the infrastructure that lets far larger flows of capital move through Bitcoin products down the road. That's institutional plumbing.
And plumbing always gets built before the flood arrives, never after. Paul Tuda Jones told a similar story during the same quarter. His firm Tudtor Investment Corporation increased its Bitcoin exposure by nearly 19%. That smart money loading up while the broader public was still heading for the exits.
All of this happened while Net Spot Bitcoin ETF flows for the year was still roughly -4.5 billion. Most people look at that negative number and conclude the rotation hasn't started yet. I'd argue the opposite. The rotation has already started. It's just happening beneath the surface in places most people never think to look.
Which brings us back to Draen Miller and the hidden bridge. Again, he didn't simply buy Bitcoin. He bought four Bitcoin mining companies. And in doing so, he positioned himself exactly where the AI boom and the Bitcoin boom physically collide. Once you understand the scarce resource both industries are competing for right now, his trade stops looking unusual and starts looking obvious in hindsight.
The bridge is the miners themselves because they sit at the intersection of three things the AI economy desperately needs. Power, comput, infrastructure, and eventually digital money. Start with power. Bitcoin miners spent the better part of the last decade hunting down the cheapest, most abundant electricity on Earth.
They secured land. They negotiated long-term contracts with utilities. They built substations. They connected massive sites directly to the grid. That is precisely the infrastructure the AI industry needs today and can't build fast enough on its own. Everyone talks about GPUs like they're the whole story, but a GPU without electricity is nothing more than an expensive paper weight.
The real constraint isn't capital. AI companies can raise billions overnight if they need to. The real constraint is time. Getting a new site connected to the grid can take anywhere from 5 to 7 years. Nobody can manufacture 5 years of lid time with a bigger check. Bitcoin miners already control the electrical on-ramps that AI companies are desperate for right now.
Today, yes, the hardware is different. Bitcoin miners run AS6 AI data centers run GPUs, but the physical foundation underneath both is identical. Land, power, grid connections, substations, and data center shells. That's one side of the bridge. Here's where the trade gets genuinely interesting. Bitcoin mining gives these companies high beta exposure to Bitcoin's price.
When Bitcoin runs, mining stocks tend to run harder, giving investors real upside talk. But the high performance computing and data center leasing side of the business gives these same companies something mining alone never could. Steady long-term contract backed revenue. Bitcoin supplies the torque.
HPC supplies the ballast. One side captures the upside, the other side stabilizes the business while everyone waits for that upside to arrive. That's effectively two separate engines bolted onto the same company. We're already watching this play out in real numbers. Bit dear mined 2694 Bitcoin in the second quarter and in that same window signed a $4.7 billion 16-year AI and HPC lease.
Riot mined another 1587 Bitcoin while also contracting data center capacity with AMD and a leading Frontier AI lab. These companies aren't choosing between Bitcoin and AI. They're monetizing the same physical footprint twice. Once through mining and once through leasing capacity to the AI buildout. I've been personally involved in Bitcoin mining since 2017 through venture positions and as an adviser to a self-hosted mining platform.
So I've seen this business from the inside, not just from a spreadsheet. And to me, this dual engine structure is exactly the kind of setup you want. High beta upside from Bitcoin, predictable income from AI infrastructure demand, and ownership of the physical power assets. both industries can't function without.
But the connection between these two booms runs even deeper than electricity because AI isn't only going to need power, it's going to need money. Mckenzie estimates that by 2030, AI agents could mediate somewhere between three and 5 trillion of global consumer commerce. These agents will discover products, compare prices, negotiate purchases, hire other agents, and pay for their own compute around the clock every single day without a human in the loop.
That kind of activity needs money that is digital by default, programmable, borderless, and able to move at any hour. I believe Bitcoin becomes that settlement asset for the machine economy, functioning as the scarce reserve, the collateral layer, and the final settlement rail, while faster layers built on top handle the actual machine to-achine payments.
Zoom all the way out and you can see what's colliding. AI is a machine for creating rapid expanding abundance. Bitcoin is a machine for enforcing absolute scarcity. AI doesn't weaken the case for Bitcoin. It strengthens it. That's the gateway. Draen Miller bought into contracted HPC income today. High beta Bitcoin upside tomorrow.
Both wrapped inside the same four companies. So the next question is whether Bitcoin itself is actually ready to move. I believe the bottom is already behind us. And I want to walk you through two indicators that support that view out of the dozens worth tracking. Start with the 200E moving average.
Bitcoin's average weekly price smoothed out over roughly four years. It currently sits around $64,000. Bitcoin briefly closed below that line and then reclaimed it almost immediately. That line isn't magic and it isn't a guaranteed floor, but historically whenever price has touched or dipped below it, that zone has marked a deep value opportunity, even if it occasionally stays cheap for a while afterward.
The second indicator matters even more because markets ultimately come down to something simple. Prices stop falling when sellers run out and prices stop rising when buyers run out. I call this the seller exhaustion constant. It combines two things. How much Bitcoin supply is currently sitting in profit and how volatile the price has been recently.
When profitability drops and volatility collapses at the same time, that tells you selling pressure is drying up. Right now, this metric sits in territory that has historically marked a bottom. In other words, sellers appear exhausted. We actually got a live stress test of this theory recently. The cold card hardware wallet hack resulted in more than $130 million worth of Bitcoin being stolen.
The kind of headline that would have crushed a fragile market a year or two ago. Instead, Bitcoin barely reacted. The market absorbed the bad news and kept moving because that's exactly how bottoms behave. Bad news stops being able to push price lower before good news is able to push it higher. Put it together and here's where I stand.
Sellers look exhausted. Large holders are accumulating and I believe the bottom is already in. That doesn't guarantee there won't be a retest. It means sellers appear to have lost control of the market. And when real demand shows up, there may not be much resistance left standing in its way. So, what should you actually do with all of this?
First, don't blindly copy Draen Miller's portfolio. And honestly, don't blindly copy anyone's portfolio. You can't borrow someone else's conviction. You have to build your own. Before you put money into anything, ask yourself five questions. What am I actually buying and why? What am I expecting to happen to it up or down?
and through what mechanism would that actually occur? Over what time frame am I expecting that to play out? And finally, what will I actually do once it happens? Once you've answered those, if you want direct exposure to Bitcoin itself, owning it outright remains the cleanest way to get that exposure.
If you've already built that core position and you're comfortable adding some risk on top, Bitcoin miners with AI infrastructure exposure can be a reasonable compliment, but treat them as different tools for different jobs inside a portfolio, not interchangeable bets. Finally, here are the three signals I'm personally watching to confirm this rotation is real and not just noise.
First, spot Bitcoin ETF flows. If year- to- date flows turn positive, that tells you institutional demand has genuinely returned. Second, Bitcoin's performance relative to semiconductor stocks. If Bitcoin keeps outperforming over multiple consecutive weeks, that's a sign market leadership is actually shifting.
Third, Draen Miller's next 13F filing. If he holds or adds to these mining positions and other major investors start showing up in the same names, that tells you the rotation is spreading beyond one early mover. I want to be clear about something important though. These are confirmation signals, not trading signals.
By the time they become obvious, you'll have more certainty, sure, but so will everyone else. And that certainty comes at a steep price, the same confirmation price we talked about earlier. At the end of the day, the real decision isn't whether you can call the exact bottom because nobody can do that with certainty.
The real decision is whether your current Bitcoin exposure and your long-term plan actually matches your own conviction about where this is headed. Do the math now before the market makes that decision for you. Let me know what you think in the comments. As always, to your success, I'm
Comments 0