월스트리트의 다음 집착은 티셔츠도 접을 수 없다(아직)

Wall Street's Next Obsession Can't Even Fold a T Shirt (Yet) THE DAILY --- H.E.A.T. Torching Wall Street’s Obsolete Playbook. Delivered straight to your inbox 5 days a week. Subscribe Home Posts Wall Street's Next Obsession Can't Even Fold a T Shirt (Yet) Wall Street's Next Obsession Can't Even Fold a T Shirt (Yet) Dec 23, 2025 I’ve been a trader and investor for 44 years. I left Wall Street long ago—-once I understood that their obsolete advice is designed to profit them, not you. Today, my firm manages around $4 billion in ETFs, and I don’t answer to anybody. I tell the truth because trying to fool investors doesn’t help them, or me. In Daily --- H.E.A.T. , I show you how to Hedge against disaster, find your Edge, exploit Asymmetric opportunities, and ride major Themes before Wall Street catches on. ❝ I’m hosting a webinar entitled “Why Covered Call ETFs Suck and What to Do Instead” (More Info Below) January 15 2-3pm. Sign Up Here Table of Contents --- H.E.A.T. The Setup: Sci‑Fi Narrative, Warehouse Reality The Hard Problems: It’s Not Just the Brain Where the Real Money Likely Flows Likely Winners: “Boring” Automation, Not Sci‑Fi Bu … Watchlist: Big Upside… but Prove‑It Risk Likely Losers (for Now): Where the Hype outruns th … Takeaways --- News vs. Noise: What’s Moving Markets Today What Actually Happened Why This Matters Beyond FX Nerds 1. Higher odds of actual yen intervention 2. The slow death of “free yen” as a funding curre … 3. BoJ’s policy dilemma = higher term premium risk 4. AI / high‑beta & “plumbing” angle Takeaways --- A Stock I’m Watching Why Covered Call ETFs Suck-And What To Do Instead --- H.E.A.T. Humanoid Robots Are Coming, As Soon As They Learn to Fold Clothes At a Silicon Valley summit, small robots roamed and poured lattes, while evangelists hailed new AI techniques as transformative. But full-size prototypes were scarce. www.bloomberg.com/news/articles/2025-12-19/humanoid-robots-are-emerging-as-the-next-big-ai-breakthrough?srnd=homepage-americas If you want to understand where the next AI mania is forming, don’t look at chatbots and image models. Look at the awkward, over‑priced machines that can’t fold a T‑shirt. Over the last 18 months, “humanoid robotics” has become the new magic phrase in Sand Hill Road pitch decks: Figure AI raising money at ~$40B valuation with backing from $MSFT, $NVDA, $AMZN‑adjacent capital and Jeff Bezos. China announcing a $138B state robotics fund and companies like Unitree selling humanoids for as little as ~$6K a unit. A dozen US and European startups (Figure, 1X, Agility, Physical Intelligence, Skild AI, etc.) promising “general‑purpose labor” in a box. And yet: at the recent Humanoids Summit in Silicon Valley, the reality looked more like a science‑fair than a labor revolution. Most of the robots on the floor weren’t sleek, human‑sized helpers. They were short, wobbly, often leashed to tables, pouring lattes or failing — in front of investors — to fold a shirt. This is exactly what a young bubble looks like: huge capital, huge promises, and hardware that’s just barely good enough to keep hope alive. The question for us isn’t “will robots be big?” (they will). It’s: what actually makes money in this cycle… and who eats the losses when the humanoid hype outruns reality? The Setup: Sci‑Fi Narrative, Warehouse Reality The sales pitch is seductive: ❝ “Large language models give robots a ‘general brain.’ Hook that brain to a humanoid body, and you get a universal worker that can do anything a person can do.” That’s why so many startups are going full “two arms, two legs”: They want robots that can operate in human‑designed spaces without re‑architecting every warehouse, kitchen, and factory. They want to sell “labor as a service,” not just a machine — bill by the hour instead of one‑time capex. They want the “PC moment” for robots: cheaper hardware, downloadable open‑source models, hobbyists hacking away in garages. Reality check from the summit: Full‑size humanoids were rare and heavily controlled. A fall isn’t just embarrassing — it’s a safety and liability event. The most common bots were ~4‑foot Unitree platforms used as software testbeds (and for “Ultimate Fighting Bots,” which couldn’t even demo because the controller batteries died). The coolest demos were narrow : a barista arm that can make lattes ~95% of the time, warehouse bots walking stairs in carefully staged videos, robots that can move boxes but struggle with soft, deformable stuff like laundry. One painful live example: a humanoid from Weave spent minutes trying and failing to fold a T‑shirt. For investors dreaming of “robot labor replacing low‑wage workers,” that’s a very on‑the‑nose metaphor. Meanwhile, the only real “in‑production” humanoid was Agility’s Digit — doing glamorous jobs like moving boxes and loading pallets in warehouses. That’s where the actual money is today: boring, repetitive, semi‑structured work in logistics, not sci‑fi butlers at home. The Hard Problems: It’s Not Just the Brain The AI crowd tends to think robotics is “just a software problem.” The roboticists at this summit know better. 1. Messy physics Popcorn gumming up sensors in movie‑theater bots. Mice chewing the wiring of waiter robots in restaurants. Slippery floors, icy sidewalks, cluttered shelves, and customers doing weird things. The “last 5%” of edge cases is where robots usually die. 2. Dexterity and data Language models learned from the internet. Robots don’t have that luxury. You have to collect data: video, sensor logs, human teleoperation traces, simulation runs. Startups like LineWise and Adamo are literally building businesses just to provide training data and remote teleoperators to “puppet” robots at scale, so the models have something to learn from. It’s the right direction — but it’s slow, expensive, and physically constrained in a way text never was. 3. Human acceptance and workplace politics Even if the robots work, you still have to deploy them: Most companies don’t have a “robot operations” team. Line workers see humanoids as job threats, not teammates; they don’t maintain them, they sabotage or ignore them. Management has to figure out new safety protocols, shift structures, maintenance budgets — none of which shows up in the sexy VC pitch deck. In other words: we are early . Think Internet 1997, not iPhone 2015. That’s good news if you’re an investor: it means the real money is still ahead — if you’re in the right part of the stack. Where the Real Money Likely Flows Forget the YouTube robot parkour videos. This is how I’d actually think about the theme. Likely Winners: “Boring” Automation, Not Sci‑Fi Butlers 1. Industrial automation incumbents – the “robot rails” These are the companies that already run factories and warehouses and quietly bolt on more robotics every year: $ROK (Rockwell Automation) – PLCs, drives, factory control systems. Robots need a control layer; Rockwell is already embedded there. $ABB , $FANUY (Fanuc) , $YASKY (Yaskawa) – industrial robots, welding arms, pick‑and‑place machines, motion control. $ETN (Eaton) , $IR , $EMR – power, pneumatics, compressors, motion components that every robotized facility depends on. These guys don’t need humanoids to work. They win as factories and warehouses keep automating, with or without sci‑fi hardware. Humanoids, if they succeed, will likely be integrated into these ecosystems , not replace them. 2. “Physical AI” in warehouses and logistics Agility Robotics (Digit) is already loading boxes in real warehouses. Other private players (Figure, 1X, etc.) are chasing the same “boring hard labor” niche: box handling, palletizing, trailer unloading, kitting. You can’t buy most of these directly (they’re private), so the public ways to play the theme are: Major warehouse owners / 3PLs who benefit from higher throughput and lower labor cost as bots scale in (and who will standardize on winning platforms). Big platforms like $AMZN , which quietly becomes the world’s largest owner/operator of industrial robots as it automates its own network. 3. Components and “robot enabling tech” Humanoids are component‑hungry : Edge compute: $NVDA, $AMD, $QCOM, $AVGO for onboard AI inference and sensor fusion. Sensors: LiDAR, depth cameras, IMUs; think $MBLY, $ON, and niche sensor specialists. Actuators/servos, batteries, power electronics: largely hidden inside diversified industrials and EV supply chains. You don’t get a pure‑play “robot servo” stock, but you can own the diversified names whose growth quietly tracks the robot build‑out instead of living or dying by it. Watchlist: Big Upside… but Prove‑It Risk 1. Humanoid pure plays (pre‑IPO and future SPAC bait) Figure AI at ~$39B private valuation with $MSFT, $NVDA, and Bezos on the cap table. 1X, Agility, Unitree, and a zoo of YC‑backed humanoid/teleop startups. These are lottery tickets for the VCs and strategics right now . For public investors, they’re future IPO/PIPE/SPAC fodder. They go on the watchlist , not the “buy” list: Unit economics are unproven. Hardware failure modes are nasty. Customer concentration risk is massive (one Amazon/Tesla/UPS decision can make or break a small vendor). When these eventually hit the public market, you’ll want to see: Payback periods under ~3–4 years at realistic utilization. Strong service / recurring revenue mix, not just one‑off hardware. Diversified customers, not one mega‑contract. Until then, note the names, don’t chase the headline hype. 2. “Robot software / data” platforms The LineWise / Adamo types — companies who: Provide teleoperators in lower‑cost geos to drive robots and create training data. Build simulation / digital‑twin platforms to train robot policies safely. Sell middleware that connects LLMs to robot control stacks. This is where you could see a “$PATH‑for‑physical‑robots” emerge — but it’s early, fragmented, and mostly private. Worth tracking as a category; tough to underwrite as an investable theme today. 3. Niche service robots Restaurant bots, hospital delivery carts, hotel concierges, co‑working greeters. These can be good local businesses but often struggle to scale globally: Highly domain‑specific. Brutally sensitive to changes in labor markets, regulations, and customer behavior. Single‑use hardware with limited reuse outside the niche. If any of these names come public with “AI robotics” plastered on the S‑1, they’ll likely trade more like cyclical equipment vendors than hyper‑growth software. Likely Losers (for Now): Where the Hype outruns the Cash Flow 1. Consumer humanoids and “home butlers” The dream: a $20K robot that cleans, cooks, and tidies. The math: Today, a ~$6K Unitree humanoid can do backflips, but not reliably handle dishes or laundry. Households are far less tolerant of failure than warehouses. Service, support, and safety‑compliance costs are huge. We may eventually get useful home bots — but as an investable theme , this is likely the last leg of the cycle, not the first. Think of it as “dot‑com pets.com phase,” not “Cisco 1995.” 2. Hype‑driven microcaps with “robotics” in the name There will be (and already are) tiny public companies that: Rebrand around “AI + robotics.” Show slick videos of off‑the‑shelf Chinese platforms with custom paint jobs. Talk about total addressable markets in the trillions. If revenue is tiny, margins are negative, and R&D is mostly “integration” work on commodity hardware, they’re not robots — they’re marketing. These are the classic “AI meme stock” candidates that spike on press releases and crash on earnings. 3. Over‑levered robot landlords The CoreWeave lesson applies to robots, too: if your business model is: ❝ Borrow expensively → buy hardware → rent it out cheaply → hope you “scale into” profit later… …you’re effectively running a leveraged carry trade on hardware costs and utilization. As soon as: Funding costs rise Utilization dips Or customers pressure pricing the equity discovers gravity. If we start seeing “robot‑as‑a‑service” companies funded like junk‑rated GPU clouds — heavy debt, low margins — they’ll be first in line when the next credit wobble hits. Takeaways 1. Humanoid robots are real… but the first big money is in warehouses, not living rooms. The robots that actually ship in volume and earn their keep will move boxes, not fold your socks. The investable angle today is industrial automation and logistics — not sci‑fi housekeepers. 2. This is a hardware boom attached to an AI brain. LLMs unlock new capabilities, but physical reality is still hard. That means longer cycles, more capex, and more ways to go bust than a pure software wave. Position higher in the stack: control systems, factories, components, power. 3. The “PC moment” for robots is starting — but it will be messy and uneven. Cheap platforms like Unitree’s sub‑$10K humanoids plus downloadable models will unleash a wave of experimentation. Most projects will fail. A few will become the “DOS/Windows” of physical automation. You want exposure to the infrastructure that all of them need, not just the one that wins. 4. As always: own the rails, not the circus. History says the people who made the consistent money in railroads weren’t the tiny frontier lines; it was the steel, sleepers, and rights‑of‑way. In this cycle, that maps to: Industrial automation incumbents Power / grid / component suppliers The big platforms that deploy robots at scale inside their own networks The humanoid robot story is going to make for amazing videos and wild private‑market valuations over the next few years. Our job is to separate the sci‑fi sizzle from the boring cash flows underneath — and make sure we’re getting paid long before a robot can finally fold our laundry. --- News vs. Noise: What’s Moving Markets Today Japan just reminded everyone that the “free yen” era is ending, and when the world’s favorite funding currency wobbles, it eventually shows up in your bond yields and risk trades. What Actually Happened The yen slid again after the BoJ hiked to 0.75% , its highest policy rate in 30 years. USD/JPY is ~157.5 , and the yen just hit an all‑time intraday low vs the euro (~ • 185) . Japan’s top FX official, Atsushi Mimura , called the move “one‑sided” and “sudden,” and said Tokyo is prepared to take “appropriate responses against excessive moves” – classic pre‑intervention language. At the same time, JGB yields are ripping higher : – 2‑year JGBs at a record ~1.1% – 10‑year JGBs >2% after breaking that level on Friday. Drivers: – Weak yen → imported inflation risk – A huge new Takaichi stimulus package that will need fresh JGB issuance → more supply, higher yields. Tokyo basically has three problems at once : a sliding currency, rising inflation risk, and a bond market that’s suddenly waking up after 30 years of sedation. Why This Matters Beyond FX Nerds 1. Higher odds of actual yen intervention Mimura’s wording (“one‑sided,” “sudden,” “excessive moves”) is step one in the classic Japanese playbook: verbal warnings → stronger language (“disorderly”) → actual USDJPY selling out of the MoF. If they pull the trigger , the impact is usually: A sharp yen spike (USDJPY down hard) Forced unwinds of yen‑funded carry trades (levered longs in EM, credit, high beta/AI, crypto) A short, violent “risk‑off” episode even if the macro backdrop hasn’t changed News vs. noise: Daily jawboning = mostly noise . Language turning to “disorderly” plus a big intraday yen spike = potential news for global risk assets. 2. The slow death of “free yen” as a funding currency The BoJ has now: Ditched negative rates Hiked to 0.75% , with markets nervously pricing the risk of more Allowed 10‑year yields to sit above 2% for the first time in decades Over the next 2–3 years , that has two structural implications: Carry trade less attractive Borrowing in yen to buy U.S. stocks, Treasuries, EM, crypto, etc. becomes more expensive and more volatile. Repatriation risk creeps higher As yen assets finally yield something, Japanese pensions/insurers eventually have more incentive to bring money home, at the margin selling some USTs and foreign bonds. HSBC and others have warned that if short‑term Japanese rates drift toward 1–2%, it could exert a “gradual but perceptible” upward pressure on global borrowing costs. News vs. noise: The regime shift is the real story: the world is losing a 30‑year anchor of ultra‑cheap yen funding, and that makes it harder to keep global long rates pinned forever. 3. BoJ’s policy dilemma = higher term premium risk Japan is now in a bind: Let the yen keep sliding → imported inflation, political pain for households, more pressure to hike. Or tighten faster (hikes or more aggressive balance‑sheet moves) → higher JGB yields, more stress on a debt‑heavy economy and banks stuffed with low‑coupon JGBs. Add Takaichi’s big stimulus (more bond supply) and you have a recipe for stickier JGB yields , not a round‑trip back to zero. For global investors, that matters because Japanese yields are one more brick in the “higher term premium” wall, alongside: U.S. deficits Fed’s fading ability to control the whole curve even with QE “lite” BOJ stepping away from being the world’s sole buyer of last resort News vs. noise: A 5–10 bps daily move in JGBs is noise. A sustained JGB 10‑year >2% with talk of more hikes is news for global duration: it raises the bar for U.S./European bonds and compresses the room for “lower for longer.” 4. AI / high‑beta & “plumbing” angle I’ve been writing a lot about: AI‑capex monsters Global liquidity The slow shift away from central‑bank omnipotence This yen story fits that arc: If BoJ + MoF are forced into real tightening or FX intervention, it adds another source of cross‑asset volatility on top of Fed QE‑lite, U.S. fiscal excess, and AI‑capex funding stress. Yen carry has historically been a big silent backer for high‑beta trades (AI, small caps, EM, crypto). A more volatile yen regime means more random “risk‑off” air pockets when those trades get unwound. It’s not a direct “sell AI” signal, but it’s another reason to favor self‑funders with real cash flow over the levered tourists that depend on perpetually easy global money. Takeaways Japan just flashed a yellow card on the yen. Mimura’s “one‑sided, sudden” language is the first real hint that Tokyo is thinking about FX intervention if USDJPY keeps grinding higher. That’s a setup for a sharp, tradable yen squeeze at some point, not a stable trend. The era of free yen funding is ending, slowly but surely. A 0.75% policy rate and 10‑year JGBs above 2% don’t kill the carry trade overnight, but they eat away at the structural tailwind that’s financed global risk for 30 years. Global term premia have another reason to drift higher. Between Japan’s rising yields, new stimulus‑driven JGB issuance, and the BoJ’s slow normalization, there’s one more force pushing against the idea of “U.S. 10‑year back to 1–2% and staying there.” Near‑term, this is more a “volatility risk” than a trend change. Day‑to‑day yen weakness with boilerplate warnings is mostly noise. The real “news” will be either (a) an actual MoF intervention spike, or (b) a clear shift in BoJ signaling toward faster hikes if weak JPY starts bleeding into domestic inflation. Portfolio angle: treat yen headlines as a plumbing tell , not a stock‑picking signal. They matter for how much hidden leverage is in carry trades, how fragile global bond markets are to repatriation flows, and how many shock absorbers are left if AI‑capex or credit has a wobble. --- A Stock I’m Watching Today’s stock is Universal Display (OLED)….. OLED (Universal Display) is a classic “picks-and-shovels” way to play the multi-year shift toward OLED screens without having to underwrite the brutal economics of being a panel maker. The company’s model is unusually high-quality for hardware: it sells proprietary phosphorescent emitter materials and collects royalties/licensing fees tied to OLED production, so it benefits as OLED penetration expands across end markets. The near-term narrative is the next leg of OLED adoption beyond smartphones—IT displays (laptops/tablets), automotive (larger, curved, higher-reliability panels), and premium TV—where power efficiency and brightness improvements matter more, and where “tandem” OLED architectures can pull forward more material content per device. The asymmetric optionality is that any meaningful progress toward commercial phosphorescent blue (or broader emitter efficiency gains) would likely be a step-function tailwind to both performance and economics across the OLED ecosystem—exactly the kind of upside that’s hard for the market to fully price in ahead of time. The key risks to watch are panel-maker capex cycles (which can be lumpy), customer concentration, and the long-run “next display tech” debate (microLED, alternative blue solutions), but OLED’s moat is its IP + deep integration with the supply chain, not a single product cycle. Net: OLED is a “quiet compounder with a call option” profile—less headline-driven than AI semis, but levered to durable secular adoption with real upside convexity if the tech roadmap hits. How Else I Can Help You Beat Wall Street at Its Own Game Inside --- H.E.A.T. is our monthly webinar series, sign up for this month’s webinar below…. Why Covered Call ETFs Suck-And What To Do Instead Thursday January 15, 2-3PM EST Covered call ETFs are everywhere — and everyone thinks they’ve found a “safe” way to collect yield in a sideways market. The truth? Most of them suck. They cap your upside, mislead investors with “yield” that’s really your own money coming back , and often trail just owning the stock by a mile. Join me for a brutally honest breakdown of how these funds actually work — and what you should be doing instead. What You’ll Learn: 🔥 Why “high yield” covered call ETFs are often just returning your own capital 📉 How most call-writing strategies quietly destroy compounding 🚫 Why owning covered calls in bull markets is like running a marathon in a weighted vest 💡 The simple structure that can fix these problems — and where the real daily income opportunities are hiding Sign Up Here All rights reserved. Keep Reading No posts found View more caret-right 0