Six Things That Actually Mattered This Week

The week looked contained from the outside.

Stocks fell, recovered, then fell again. Oil held above $100. The Fed held rates. Nothing announced itself as a turning point.

But underneath the surface, six distinct shifts moved in the same direction. Each looked like a separate story on the day it broke. By Friday they were all saying the same thing. The assumptions that priced every major asset class for the past two years are being revised. Some gradually. Some all at once.

The investors who noticed early had time to adjust. The ones still running last year's models are carrying risk that isn't showing up in their numbers yet.

I've worked for the CIA. Met four US presidents.

But meeting Elon Musk face-to-face changed everything.

My research now leads me to believe he'll announce the SpaceX IPO on March 26, 2026.

I'm sharing an "access code" to grab a pre-IPO stake. But only with the first 500 people today.

Click Here to See How to Secure Your "SpaceX Access Code"

The week opened with markets still treating Hormuz as a temporary disruption. A shipping problem. Something that resolves in weeks.

By Friday that framing was gone.

S&P Global's Karim Fawaz, whose firm advises most major energy companies, said the window to restore normal flows closed in the first week of the conflict. Kuwait's state oil CEO said production recovery takes months even after the war ends. Qatar's Ras Laffan complex needs five years of repairs. Between 10 and 20 percent of global oil and exportable natural gas are offline.

J.P. Morgan published a sequenced map showing when each region feels the impact. Asia is already in it. Africa hits early April. Europe mid-April. The U.S. around April 15. The chaos has a schedule.

By Friday, futures markets were pricing a Fed rate hike at 52% probability. One month ago they were pricing two cuts. That reversal happened because $110 oil, import prices at their highest monthly jump since March 2022, and inflation expectations rising to 3.8% collectively told the market the energy shock isn't fading. It's embedding.

Investor Takeaway 

Price events reverse when supply comes back. Supply restructuring events create new baselines. The investors still modeling a reversion to February prices are answering the wrong question. The new question is what the world looks like when this becomes the baseline.

The week made clear that the AI buildout has constraints capital alone cannot solve.

Broadcom said TSMC is hitting production limits that don't expand meaningfully until 2027. PCB lead times stretched from six weeks to six months. Laser components face the same pressure. Customers are signing four-year supply agreements because spot availability is already unreliable and getting worse.

OpenAI, which spent a year signaling it would build its own data centers, stepped back. Build delays, supply friction, and financing gaps exposed the gap between AI ambition and infrastructure reality. The company shifted from builder to buyer, securing capacity from Microsoft, Google, and Amazon. That means OpenAI's compute costs are now someone else's revenue permanently.

A different ceiling emerged on the grid side. Constellation restarted Three Mile Island to power Microsoft data centers by 2027. PJM, the grid operator covering the largest data center market in the world, told Constellation the transmission upgrades won't be ready until 2031. Four years late. The plant will be ready. The wires won't.

Investor Takeaway 

AI demand is real. The physical infrastructure required to serve it is constrained in ways that don't resolve with more funding. Any AI infrastructure investment where supply or grid connection is assumed rather than confirmed is carrying timeline risk that doesn't show up in the revenue model.

That’s what the media wants you to focus on.

But behind the scenes, it seems President Trump is quietly preparing something far more shocking — that will leave even his most loyal MAGA patriots stunned. 

It’s NOT being debated on cable news or on X.

But it could make you enormously rich in 2026. Click here to find out why immediately.

Private credit entered the week under pressure. It left under considerably more.

Apollo gated its $25 billion fund at 5% after receiving redemption requests for 11%. FS KKR was cut to junk by Moody's with a non-accrual rate of 5.5%. Stone Ridge paid 11 cents on the dollar for redemption requests. Ares and BlackRock restricted withdrawals. A California pension fund canceled a $140 million private credit allocation, citing lower returns and the hassle of the asset class.

JPMorgan filed for a new private credit fund offering 7.5% quarterly redemptions, 50% more than the funds gating investors. Either JPMorgan built something structurally more liquid or it's pricing in risk the gating funds won't acknowledge. Capital flows at launch will tell you which.

Lloyd Blankfein, who steered Goldman through 2008, told CNBC that private market assets haven't been tested against real prices. "There has to be a reckoning," he said. "We haven't had one, and the longer between reckonings, the worse it could potentially be."

Investor Takeaway 

Private credit's stability assumption was built on steady inflows and patient investors. Both are changing at the same time. The marks haven't moved. The exit behavior already has. Watch non-accruals, PIK income, and gate announcements. Those numbers move before the official story changes.

Thursday produced three separate rule changes that reset baselines models were built on.

A federal judge ruled the Pentagon's designation of Anthropic as a supply chain risk was First Amendment retaliation. The government's own records showed the ban was triggered by Anthropic taking a contract dispute public. The judge called it classic illegal retaliation. Every AI firm negotiating government contracts now has a precedent to cite. The company that lost hundreds of millions in canceled contracts Monday was IPO-ready by Thursday because the ban was illegal.

Intercontinental Exchange invested $600 million in Polymarket the same week a bipartisan Senate bill attempted to ban sports contracts from prediction markets. ICE owns the NYSE. It is not making a $2 billion bet on sports contracts. It is betting that prediction markets become institutional derivatives infrastructure. Insurance companies hedging hurricane exposure. Asset managers expressing macro views. That business isn't what the Senate bill targets.

And the probability of a Fed rate hike by year-end crossed 50% for the first time, flipping a market that was pricing cuts just weeks ago.

Investor Takeaway 

Three baselines reset in one week. Anthropic's injunction, Constellation's grid delay, and the rate hike probability crossing 50% all changed the terms on which positions were built. The investors who noticed early had time to adjust.

Why choose between growth or income when you can have both?

Our new report reveals 10 “Double Engine” stocks — companies built for rising dividends and breakout price gains.

Each has the scale, cash flow, and catalysts to outperform as markets rotate after the Fed’s pivot.

These are portfolio workhorses — reliable payouts today, compounding gains tomorrow.

Access the Top 10 Best Stocks to Own in 2026 — free for a limited time >>

The week's most instructive capital allocation stories didn't involve model companies or headline acquisitions.

KKR bought a majority stake in CoolIT Systems, a data center cooling company, in 2023 at a $270 million valuation. Nobody wrote about it. This week KKR agreed to sell CoolIT to Ecolab for $4.75 billion. Fifteen times the equity in two years. Revenue quadrupled. The company's 640 employees will receive average payouts of $240,000.

The mechanism was simple. AI data centers generate heat. Liquid cooling is the most efficient solution at scale. CoolIT was already in the right part of the stack when the buildout accelerated.

Delta Air Lines bought an aging Philadelphia oil refinery in 2012. Everyone thought it was strange. This week, with jet fuel at $192 per barrel against $100 crude, Delta is paying the crack spread to itself while United faces a potential $11 billion increase in annual fuel costs. The decision that looked overcautious in 2012 is the only hedge that works in 2026.

Investor Takeaway 

The AI cycle's returns are concentrating in the physical layer solving the constraints the models depend on. Power. Cooling. Chips. The decisions that looked strange in calm conditions are the ones that work when conditions aren't. The edge isn't predicting the stress. It's owning what breaks when it shows up.

The least discussed story of the week may carry the longest tail.

Corn acres are expected to fall 4.5% this year. Spring wheat plantings are on track for their lowest level since 1970. Farmers are switching to soybeans because soybeans need less nitrogen fertilizer. Urea is up 40% since the war started. Anhydrous ammonia up nearly 20%.

The USDA releases its annual planting intentions report Tuesday. Analysts are already warning the survey predates the war's full cost impact. The report will understate what's actually happening in fields right now.

Lower corn and wheat acreage means lower yields in fall. Lower yields tighten food supply. Tighter food supply pushes grocery prices higher. That shows up in CPI months after the planting decision. The fertilizer shock is already embedded in the harvest. It just hasn't landed in the data yet.

Investor Takeaway 

The Iran war's inflation consequences are arriving in stages. Energy hit first. Fertilizer is hitting now. Food prices hit at harvest. Each stage arrives months after the decision that caused it. The CPI reports that looked calm last month were measuring a world that no longer exists.

OpenAI is gearing up for a historic IPO, and Silicon Valley insider Luke Lango has found a way for you to invest BEFORE the announcement is even made. 

You don't need to file any special paperwork… buy shares from a former employee… have a source on the inside – or jump through any other hurdles.

Best of all, all you need is just $10 to get started.

Six sequences. One consistent signal.

The world that priced markets for the past two years changed this week. Energy is restructuring, not disrupting. AI hit physical limits that money can't solve on the market's preferred timeline. Private credit's test widened from one fund to an entire asset class. Three rules reset in 72 hours. The boring bets paid off better than the narrative ones. And the farming calendar was rewritten in a way that won't show up in the data for months.

None of these stories announced themselves as turning points.

That's usually how the important weeks work. The question heading into next week isn't whether things changed. They did. It's which of your positions still reflects the old version.