ORCL was the wrecking ball yesterday; today we’re green again, but the message didn’t change: the market just learned (again) that AI isn’t a high-margin software business—it's a capital project. The Information said Oracle did ~$900M of NVDA‐GPU rentals last quarter at 14% gross margin (retail-store margins), and even lost ~$100M on Blackwell in the same period. Pushback helped ORCL halve its losses into the close, but the signal is clear: when you rent very expensive chips into a price-competitive market, the P&L gets thin—fast.

Here was the key point…

Rhyming history: this is why gray-hairs get twitchy around data centers. The last time the street thought “build it all now” was Exodus in 2000–01—great long-term demand, fatal near-term leverage. Even Oracle is pivoting. A year ago Ellison was hyped about a gigawatt nuclear-powered DC; last month Safra Catz reminded everyone: “we don’t own the land or the buildings”—they’re shifting to revenue-gear, not concrete. That’s the tell: when the economics don’t pencil, the smart money de-risks the footprint.

Rhyming history ≠ destiny. Michael Dell says he doesn’t see overbuild yet. Fair. These are multi-year contracts and the flywheel can run longer than bears think. Just remember: more capacity rarely brings pricing power……

This echoes my thoughts, at some point this bubble pops, but not yet.

Meanwhile, the debasement trade keeps doing laps in the background. Gold > $4,000 and still climbing as Japan elects a fiscal/monetary dove (yen down, JGB 30-year up) and Europe wrestles with deficits and politics. Griffin said the quiet part out loud: gold is becoming the safe harbor the dollar used to be. This isn’t just a U.S. story—fiat faith is leaking globally. That’s why gold and crypto can rally with AI and equities: investors are betting on growth and hedging policy at the same time.

AI financing is getting more… creative. Musk’s xAI is reportedly structuring a $20B SPV to buy NVDA GPUs and rent them back to xAI—with NVDA potentially investing in the vehicle. Call it circular, call it ecosystem building—either way it extends the cycle and keeps the builders paid. But circularity cuts both ways when economics tighten.

What to do with this
• Own the cash-flow stack that gets paid whether AI margins are 70% or 14%: power + grid + thermal + interconnect. Our spine hasn’t changed—$ETN $VRT $HUBB $POWL (electrical/thermal), $CEG $VST (dispatchable power), $EQIX $DLR (interconnect). Chips/fabrics leadership stays $NVDA $AVGO $ANET; $AMD just got more real, but let the dust settle after the surge.

• Trade the froth, don’t marry it. Unprofitable “AI-halo” names (hydrogen/quantum/EV charging) work until they don’t. Tight leashes, defined exits.

• Respect the debasement bid. Keep a core in gold and the real-asset complex; the macro tail—fiscal dominance risk, global politics, curve steepening—still blows that sail.

• Hedge the middle. Use put spreads/ratio puts on high-beta into headline windows; pairs help (long AI power vs. short a basket of unprofitable AI platforms). If we’re wrong and utilization ramps, we’ll know quickly via cloud gross margin and non-commenced lease burn-down.

Thursday October 23rd 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.

🔥 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

TL;DR: TD Cowen’s new Ahead of the Curve report lays out a rare consensus: Western nuclear power is not just coming back — it’s becoming the backbone of the AI + Energy industrial complex. The drivers are clear: hyperscaler demand, national security policy, and power scarcity.
But there are bottlenecks — uranium, enrichment, and permitting — that could shape who wins this race and who gets left behind.

After decades of stagnation, nuclear energy is at the center of a global rearmament of power capacity. TD Cowen’s analysts frame it as a new growth cycle fueled by two catalysts:

This combination turns nuclear from a slow, overregulated legacy sector into a high-growth, defense-grade infrastructure theme.

TD Cowen expects the Western nuclear ecosystem to expand 2–3x by 2035, and highlights that valuations of key nuclear stocks already price in far higher growth than government forecasts — meaning this is one of the few areas where the Street may still be underestimating how big it can get.

TD Cowen structures the opportunity set around four key debates investors must understand:

Perhaps the most explosive takeaway: Nuclear energy is now a data center strategy.
TD Cowen maps over 6.9GW of new nuclear-linked power deals with hyperscalers:

💥 Implication: The AI arms race is now a power arms race, and nuclear is the only clean, 24/7 baseload option that scales.

This report confirms one of the most powerful structural shifts in the global economy:
The AI-power-nuclear flywheel.

It links your top themes — AI infrastructure, power grid investment, defense-industrial rearmament — into one trade.

For timing: TD Cowen sees front-loaded policy impact through 2026–2028, with fuel-cycle scarcity sustaining the bull case well into the 2030s.

Nuclear is no longer an ESG liability — it’s a national security necessity and a tech enabler.

The U.S. is rebuilding what it dismantled:

If the 2010s were the decade of “renewables beta,” the 2020s are shaping up as the decade of atomic alpha.

Bottom Line:
TD Cowen’s “Great Nuclear Debate” makes one thing clear — this isn’t your grandfather’s nuclear cycle. It’s an AI-fueled, defense-backed, policy-protected energy supercycle.
And for investors, that’s the kind of asymmetric setup that defines generational trades.

Today’s stock is a sector and it’s a short…..

Rates are coming down so homebuilders should be doing better, they aren’t. My guess is it’s because the average consumer has no money. We are seeing similar moves in other areas. I think it’s prudent to have some hedges around a weak consumer.

The short side is much harder than the downside, you can’t hold a short as long and you need to be watching the chart. I’m traveling for the next two weeks so it’s unlikely I’m going to do much shorting. I did buy more puts on ARKK on Monday, nice timing by me

But if those stocks turn then my puts get crushed.