We’ve talked in recent weeks about the “Venezuela model” as the ultimate outcome for Iran.

Take the Iranian oil.  Sell it at market rates. The money goes into a U.S. Treasury-controlled account — removing the revenue lifeblood from the regime, reserving the money for reconstruction, and the U.S. controlling key global oil supply in the process.

Rubio detailed the mechanism at the cabinet meeting a couple of weeks ago. Then both Rubio and Bessent described their half of the architecture, in their respective congressional testimonies last week.

Today Trump said it explicitly. He said in the not too distant future, the U.S. would take Kharg Island (the terminal behind roughly 90% of Iran’s crude exports) and “assume total control of their oil and gas markets, much like we have with Venezuela.”

The oil can flow, but not on the old terms, and not back to the old hands.

So, a deal, in this framework, doesn’t mean withdrawal. It means control. 

And with that, this afternoon, Trump announced (yet another) “deal” is near …

 

This cancellation of more strikes resulted in yet another relief episode for markets — another sentiment massage (stocks up, yields down, oil down).  

Notice within this post, the coalition of countries Trump named: the United States, Israel, Saudi Arabia, UAE, Qatar, Turkey, Pakistan, Bahrain, Kuwait, Jordan, Egypt.

Notice Europe wasn’t mentioned. The energy shock that just drove the ECB to a rate hike this morning is being negotiated without their participation.

With that, let’s revisit this excerpt from my May 19 note…

Scott Bessent was in Paris, delivering the keynote at the No Money for Terror conference. He told the room of global policymakers and financiers this:

 

“The United States is hardly alone in facing the scourge of terrorism, especially from Iran. Yet, too often, we seem to be alone in our resolve to thwart it.” 

 

He then called on Europe, by name, to take action and expose Iran’s financing networks — unmask shell and front companies, shutter bank branches, dismantle proxies.

 

And he called on partners in the Middle East and Asia to root out Iran’s shadow banking networks (the China-Iran financial pipeline).

 

But this “alone in our resolve” statement was clearly aimed at Europe, along with “no room for excuses.”

 

We’ve talked about the Trump plan to restructure global trade and realign the world (away from China, back toward the U.S.), anchored by “burden sharing” — where allies and trading partners pay for access to safety (U.S. security guarantees), stability (the dollar and U.S. capital markets), and markets (U.S. consumers).

 

These are all no longer automatic, but conditional — conditional based upon alignment. 

 

And Bessent is making it clear that alignment includes: designate, expose, shutter, and dismantle Iran’s financing.

 

Or, stay ambiguous, as Europe has to this point, and discover what “conditional” looks like in practice.

 

And we’ve talked about the levers that Trump, Bessent and the new Fed Chair can pull to force political alignment — through consequence, not argument.

With this, “force political alignment through consequence” tactic in mind:  If a “deal” on Iran is coming, it still wouldn’t resolve the energy shock for Europe soon enough. Qatari LNG wouldn’t normalize for years. And the ECB’s own staff admitted last week that energy pass-through to utility bills takes “several months.”

And with Kharg on the table, relief for Europe runs through Washington. U.S. LNG and crude, Iranian barrels through a Treasury architecture, Qatari volumes returning.

Fair to say, relief comes with alignment. 

Don’t forget, Europe sits on the Trump-imposed July 4th deadline for a trade deal.

This morning we got May CPI. Hot, as expected: 4.2% on the headline, up from 3.8%.

As we discussed in Monday’s note, with the effective fed funds rate sitting around 3.62%, real rates just went more negative — now roughly minus 0.6%. The deepest negative real rate since 2023.

Now, here’s the chart of the day …

A significant technical trendline in gold broke today. This is the trendline that defined the runup over the past 17 months of the Trump presidency.

The textbook says that shouldn’t happen. Gold loves negative real rates. The deeper they go, the better it tends to do — because the policy intent behind negative real rates is typically to promote risk taking (move people out of bonds and into of higher-return assets, to overcome the drag of inflation against purchasing power). Gold tends to go up as an inflation hedge (a contra-dollar/contra-paper currency).

That said, today real rates went deeper negative than they’ve been in three years, and yet gold broke support, continuing its slide — now down close to 30% from the highs.

There are two reasonable explanations: 1) The rate market is pricing about 70% odds of a Fed hike by year-end, so gold may be trading the expectation that today’s negative real rates won’t be around for long.

And 2) the war uncertainty premium may be subsiding — the parabolic runup in gold from September was, in large part, a war bid, and this week the whole complex is unwinding together: silver down double digits, platinum down 8%, crude down 7%, dollar up.

Both sound reasonable. But if we step back, there is also a bigger story the chart may be telling.

Gold traded below $1,000 coming out of the Global Financial Crisis. It peaked this year above $5,000. We should acknowledge that the five-fold run in gold maps against fifteen years of central bank balance sheet expansion (QE, deficit underwriting, the Fed absorbing the Treasury market).

And remember, Kevin Warsh, under oath in April, called it what it is: “fiscal policy in disguise.”

An inflation in asset prices, including gold, defined that era. It wasn’t a rates trade. It was a quantity of money trade.

They printed more money, the nominal price of stuff went up. Gold, the asset with a history of offering protection against such debasement of the currency, among them.  

That all makes today interesting.

Warsh chairs his first Fed meeting in seven days.

His stated direction for policy (the one we walked through Monday) is to end the “disguise.” It’s to extract the Fed from fiscal policy, shrink the balance sheet, and pair it with lower rates.

If that regime change is real. If it’s going to happen. Then an engine of the gold bull is getting turned off. And under that lens, gold can fall while rates fall.

So if the regime story is right, gold is falling on anticipation, pricing Warsh’s credibility to do what he says he’s going to do. But, to be sure, there are a lot of obstacles (and likely internal opposition at the Fed) to work through in order to get the Fed regime change (policy and cultural) underway. 

June 17 is the first test. 

In my February 18 note, we called the race for AI supremacy a two-horse race, winner-take-all, and “very, very tight.”

Four months later, that statement still holds.

Today Anthropic released Fable 5 to the public. This is the guardrailed version of Mythos. 

Remember, Mythos is the Anthropic model we discussed in April that found and exploited decades-old security flaws on its own. It was deemed so capable they didn’t release it, and instead took it to the government and convened industry leaders to start building protection against it.

That said, around 1pm this afternoon, they announced the release of Fable 5 as the “safe” version of Mythos.

Stocks bottomed right around that time, after what was becoming another messy unwinding day for AI stocks.

So now we have another big leap forward in model capabilities — maybe the most consequential since the early February models that brought the explosion in agentic AI.

With that, the read over the past half year or so has been that — Anthropic is king, AGI is near (if not here), the U.S. has won the AI race. 

But this chart says otherwise. 

This tracks the “Intelligence Index” of the world’s top AI models (the dots).  Notice, the U.S. models are on top. But what’s clear, is that the dots are clustering tighter and tighter as the model intelligence is rising.

The frontier isn’t separating, it’s converging. OpenAI, Anthropic, Google, xAI are bunched together at the top. And the Chinese labs –DeepSeek, Alibaba, Kimi have compressed the gap, not widened it.

What’s the finish line? 

Human-level intelligence, that becomes self-improving.

The first to that finish line should be the winner, and therefore, should be in position to set standards, attract talent, and determine what technology gets embedded into governments and critical infrastructure.

And as we’ve said before, that will mean the difference between AI that serves humanity, or AI that controls humanity (serving the interest of the Chinese Communist Party).

Is Mythos (and today’s publicly available model, Fable) the model that reaches the finish line?

Maybe. Consider this: Anthropic preceded today’s model release with a note last Thursday titled, When AI builds itself (see it here).

And Anthropic’s founder Dario Amodei had already described Mythos a few months ago as “by far the most powerful AI we’ve ever built.” Around that time, he also did a podcast where he said, the surprising thing isn’t how fast they’re accelerating, it’s “how close we are to the end of the exponential.”

Not the start. The end. He’s saying the steep part of the climb is nearly behind us.

If he’s right, the convergence on that chart above may not be a pause before someone breaks away. It may be the field arriving near the finish line together.

So, still a tight race.

Which is why China is a significant threat. One Chinese model leap doesn’t have far to travel.

 

We get the May jobs report tomorrow.

For the better part of the past two years, we watched this report for “cracks in the labor market” which was the explicit condition named by the Fed — it would force the Fed’s hand.

Weak jobs meant rate cuts.

So, markets were on high alert the first Friday of every month.

That era/regime is largely behind us.

The shift came back in December. Remember, the jobs report due that month was postponed by a government shutdown. When it was finally reported mid-month, it showed the unemployment rate jumping to 4.6%.

It was the highest level in four years, and yet the market barely flinched.

Why?  The Fed had already pivoted.

The Fed cut rates for a third consecutive time earlier that month, and had turned the liquidity spigot back on (returned to expanding the balance sheet).

The conditionality of the jobs report was supplanted by signs of financial instability (some stress in the money markets) and cooling inflation.

That brings us to tomorrow’s report. 

First, rates. The story now isn’t whether a soft print triggers a rate cut. In fact, at the moment, the market is pricing in the chance of a rate hike by year-end. 

But what matters now, under a Warsh-led Fed, is getting the Fed out of the way of an economy that has pro-growth policy behind it and a technology revolution running through it. It’s a structural tailwind.

Second, the jobs number itself is communicating a different message. Under the Trump administration, government headcount is being cut while private hiring is coming back (chart below). 

 

So, for the headline payroll number, a “soft” headline could actually be a healthy rotation: fewer government jobs, more productive private ones. The composition matters maybe more than the total.

The old question was “will this scare the Fed into cutting?” The better question now, and in the coming months: “is the private economy growing, and is AI-driven productive work showing up in companies, that leads to more hiring, to drive more productive work …?”

We talked yesterday about Jensen Huang’s presentation showing the production of software going vertical over the past few months — the “$9 trillion of productivity from $3 trillion of salaries.”

And we ended by looking at the San Francisco Fed’s report on total factor productivity, which decelerated in the first quarter, to 0.60%.

Let’s talk about that gap.

Jensen focused on AI usage.  For the past year, “token consumption” has been celebrated in Silicon Valley. Big companies (particularly big tech) took a clear strategic position: press the gas pedal on AI to the floor (i.e. burn tokens). 

They turned usage into a scoreboard. Companies like Uber, Meta and Amazon ranked engineering teams by token consumption (the more, the more celebrated).

What’s the problem? A high token count doesn’t tell you a lot got done. It tells you a lot got used.

When you reward people for consumption, the meter just runs.  But you don’t necessarily solve problems, ship products, reduce costs, or create value.

And the tell, at this point, is in the recent behaviors of these same companies — they are cutting back on, or setting limits, on usage. 

The Fed’s Q1 number on total factor productivity reveals that story (it decelerated, not accelerated). It doesn’t confirm a productivity boom, at least yet. 

But it’s early. We’ve seen the investment. We’ve seen hesitation, at the enterprise level. Now we’ve seen aggressive adoption and experimentation. We’re seeing some waste. Now maybe some refinement, discipline and productivity gains.

With all of the above in mind, we get the May jobs report on Friday. Is the AI building phase increasing labor demand, or is AI visibly replacing labor? 

Is AI proving “useful” as Jensen positioned it? Or is it just getting used a lot? 

Another weekend, another deal that was supposed to be close.

Going into it, the framework was reported as all but done. Trump said the deal to reopen the Strait of Hormuz was close.

The market heard “close” and treated it as done. Oil eased. Stocks lifted.

And again, nothing.

We’ve seen this plenty of times now over the past three months.

A conciliatory headline. A market that wants to believe the pressure is over.

The reality on the ground remains: an energy supply shock. 

So what actually happened?

Trump’s terms are inflexible. He wants Iran’s enriched uranium dug up and destroyed, the Strait open immediately, and Iran agreeing it will never have a nuclear weapon.

That said, we already know these conditions. And it’s safe to assume that anyone on the other side of the negotiation with ties to the old regime would never agree to them. 

Indeed, a senior Iranian official told Reuters they never agreed to hand over the uranium at all. 

The shooting resumed. The deal fizzled again — because the terms are about ceding control of all leverage (oil and nuclear).

As we’ve discussed, this is the Venezuela model.

The oil can flow, but not on the old terms, and not back to the old hands. A deal, in this framework, doesn’t mean withdrawal. It means control.

Let’s talk about the Fed.

On Sunday night, Jerome Powell accepted a “Profile in Courage” award. In his acceptance speech, he “warned” that if an administration removes Fed officials over policy, the Fed loses its most priceless asset, its credibility.

Let’s revisit what “independence” looked like under Jay Powell. 

For the better part of two decades, the Fed has been anything but independent from fiscal policy. It financed Capitol Hill spending to the tune of trillions of dollars, through QE. Kevin Warsh (the new Fed Chair) has a name for it — “fiscal policy in disguise.”

And the Powell Fed was anything but neutral.

Remember, in Trump’s first term, Jerome Powell mechanically raised rates into a low inflation, recovering economy. Not only was it a headwind to the economy, it induced a liquidity shock.  

Then, Powell called inflation “transitory” in 2021. And he held rates at zero while prices ran at a four-decade high pace. It was the transitory lie that gave the Democrat-led Congress the cover to push for trillions more in spending (to fund the climate and social agenda).

So, the “independent” Powell-led central bank was, in practice, the financing arm of the government.

What Powell is now defending, is the Bernanke/Yellen/Powell Fed regime, under the guise of “independence.”

Warsh was hired to end that old regime — to restore independence, by shrinking the balance sheet, by ending QE as a permanent tool, and by ending the Fed’s financing of the deficit.

The boom loop we’ve been discussing for the past few months continues to materialize in corporate earnings.

More compute. More revenue. More profit. More compute (…). 

We’ve seen it in the earnings of chip and memory stocks. 

Now, it’s broadening. 

Let’s step through the AI stack.

Start with the chips.

As we’ve discussed in these notes, Nvidia put up another huge quarter — almost $82 billion of revenue in a single quarter, more than ten times its revenue of three years ago (and guided to $91 billion for the next). Its CFO told the market AI infrastructure spending is now on track to reach three to four trillion dollars a year by the end of the decade. 

Those chips have to be built.

Taiwan Semiconductor, which manufactures the most advanced ones in the world, reported a 20% jump in revenue in a single month and greenlit $45 billion in new capacity.

The chips can’t compute without memory.

Micron posted $23.9 billion in its most recent quarter, up 196% from a year earlier, and guided to a record $33.5 billion quarter at margins near 80%. Moreover, Micron has already sold out its entire 2026 production of these chips, under binding contracts. This week, the valuation crossed $1 trillion.

The chips also produce oceans of data that needs to be stored.

SanDisk reported $3 billion in a quarter and then followed it with a $5.9 billion quarter — a near doubling of revenue in ninety days.

Then the data has to be organized and put to work.

This week, Snowflake (the data-cloud company) reported product revenue of $1.33 billion, up 34%, the strongest quarterly dollar growth in its history. It raised full-year guidance, signed a $6 billion infrastructure deal with Amazon, and told us the number of customers using its AI tools jumped from 9,100 to 13,600 in a single quarter. The stock rose 37% in a day.

The AI applications need a database to run on.

MongoDB just reported revenue up 25%, its cloud database growing 29%, as it positions itself as the working memory for the coming wave of AI agents. It raised guidance too. The stock trading up as much as 36%. 

And all of it has to run on servers.

Today, Dell reported an AI server backlog of $51 billion, after AI-server revenue grew 757% compared to the same quarter last year. It raised its target for this year toward $60 billion. The stock rose 40%.

So, while the daily headlines argue tenths of a tick in inflation and the price of oil, every single layer of the AI stack is telling this story: accelerating demand, raised guidance, and not enough supply.

This is the boom loop we’ve been describing, broadening across the entire AI stack — which should broaden across the entire economy.

As we’ve discussed over the past three years, AI is indeed making the pie bigger. The trillion-dollar valuation club is indeed growing.

In a world racing toward abundance, you want to own the scarce things that abundance can’t exist without.

That’s what our two portfolios are built around. Our AI-Innovation Portfolio owns the scarce physical inputs the buildout cannot exist without — the power, the fiber, the connectors, the chip equipment, the network edge. Our Billionaire’s Portfolio owns the still-undervalued producers of the hard assets that feed it — the copper, the gold, the oil and gas, and the legacy technology being revalued as AI-critical infrastructure.

Two portfolios driven by one thesis. If you’re not yet a member, it’s a good time to get positioned for what the stack just told us is coming.

Learn more here. 

 

 

Yesterday, we talked about Trump’s Iran strategy, and how he has repeatedly referenced the “Venezuela model“– take control of the oil, fund the country and the restructuring, prevent the re-emergence of the old regime.

With that in mind, Trump held a cabinet meeting this morning. And in the middle of it, he turned to Marco Rubio and asked, “Cuba and Venezuela, what’s going on there?”

Rubio said Venezuela is in “a three-phase process … stabilization, recovery, and transition.”

Then he walked through the mechanism.

Since January 3, over 10 million barrels of Venezuelan oil have been delivered to the United States. He said, the industry is being “professionalized for the first time ever.”

The crude is sold “in the market at market rates.” And “the money is going to an account in the United States controlled and monitored by Treasury, audited by KPMG. And it’s for the first time ever, the money’s not being stolen. It’s going to the benefit of the Venezuelan people.”

That’s the Venezuela model stated by the Secretary of State. A Treasury-controlled account. KPMG audit. Market sales. Revenue restructured away from the regime, toward the people.

As we discussed in my note yesterday, this model applied to Iran would have to involve controlling Kharg Island (the terminal that moves the overwhelming majority of Iran’s crude). 

Is that why Bessent has said he expects crude prices to be “lower than pre-conflict levels when this ends” (his exact words today)? 

The U.S. Department of Treasury already controls Iran’s money. The U.S. Department of War controls the travel of oil. And with that, there’s no need for a big kinetic seizure. Time does the work (as Trump has said, “time is on our side”). 

Trump wants cheap gas. And cheap gas requires controlling the swing capacity and the chokepoints. That means Iran and Venezuela under the same architecture, with the Saudis and Abraham Accords members under the “aligned partner” architecture already negotiated.  

 

Another weekend, another deal that was supposed to be close.

The anticipation built across the weekend included “talks,” “good signs,” a “framework” near.

And again, nothing.

Instead, the U.S. struck Iranian sites, and Iran is now calling those strikes a violation of the ceasefire. We’ve seen this movie plenty of times now over the past three months.

Remember, back in my May 11 note (here), we talked about the Netanyahu interview with 60 Minutes. He said the war isn’t over — “it’s not over because there’s still nuclear material, enriched uranium that has to be taken out of Iran.”

Trump reiterated that last night, posting that the enriched uranium would need to be turned over and destroyed, or destroyed in place under inspection.

Until it happens, we should view the conciliatory headlines as theater, and the strikes as substance.

And the other key factor: oil.

Oil is the tool, the leverage, the revenue that funds the regime, the proxies and the weaponry. It’s the lever Iran has always used to manufacture instability in the global economy.

This is why “a deal” seems unlikely. Trump has pointed, repeatedly, to Venezuela as the model. And in Venezuela the U.S. effectively controls the oil — to fund the country and restructuring, and to prevent the re-emergence of the old regime.

For Iran, this model would include taking control of Kharg Island (the terminal that moves the overwhelming majority of Iran’s crude). And it’s hard to imagine the current Iranian counterparts coming to any agreement like that.

For now, time does the work. The export machine continues to choke and revenue collapses. 

Let’s talk about the Fed.

Kevin Warsh is on the job, sworn in Friday.

Warsh has said, repeatedly, that the Fed should talk less — less forward guidance, less telegraphing, fewer officials shaping markets with words.

Yet, what does the Fed calendar show for Warsh’s first week? A long line of Fed governors with public speaking engagements. The leadership has changed. The question is, how entrenched is the culture?

We may get some signals on that this week. 

Still, as we discussed last week after Nvidia’s earnings, the largest industrial buildout in modern history is underway. It’s funded, profitable and accelerating. Against that tsunami, the day-to-day market noise looks like ripples on the surface.

We saw it again today — another new record high for stocks, led by AI infrastructure, and the scarce physical inputs that feed it.

We have 25 stocks in our AI-Innovation Portfolio, focused on the companies powering this AI infrastructure buildout — for what Elon has called a coming economic “tsunami.” We have been positioned for it since June of 2023, but the buildout remains in the early stages. If you haven’t joined us yet, you can do so here …