The "Morning Squawk" crowd is high on the fumes of trillion-dollar projections and the dopamine hit of instant gratification. They look at Nvidia’s climbing valuation and see a perpetual motion machine. They look at Amazon’s shrinking delivery windows and see "progress."
They are wrong.
What we are witnessing isn't a tech renaissance. It is a capital expenditure bubble masquerading as a revolution, paired with a logistics arms race that has no profitable finish line. If you’re following the herd into these "guaranteed" winners, you aren't an investor. You're a bag holder in waiting.
The Nvidia Mirage: Why Chips Aren't the New Oil
The consensus says AI is the new industrial revolution and Nvidia is the only one selling the steam engines. This is a fundamental misunderstanding of how commodity cycles work.
In every previous tech cycle—fiber optics in the late 90s, mobile hardware in the 2010s—the initial infrastructure build-out creates a massive spike in demand that everyone mistakes for a permanent plateau. It isn't. It’s a mountain peak.
Nvidia’s $1T+ projections assume that every company on earth will continue to spend billions on H100s and Blackwell chips indefinitely. But here is the reality: The ROI on AI software is currently abysmal. I have sat in boardrooms where Fortune 500 CEOs are sweating because they’ve spent $200 million on compute and haven't seen a single percentage point of margin expansion. Eventually, the CFOs wake up. They will realize that renting or buying massive GPU clusters to summarize emails is like buying a Ferrari to deliver mail.
When that realization hits, the "extraordinary demand" vanishes. We will move from a supply-constrained market to a massive overcapacity glut. Nvidia isn't the new oil; it’s the new Cisco. Remember Cisco in 2000? They were the "essential" backbone of the internet. Their stock never recovered to those highs because the world eventually had enough routers. We are reaching the point where the world has enough "intelligence" hardware but zero idea how to make it pay the rent.
The One-Hour Delivery Death Spiral
The "Squawk" analysts love talking about Amazon’s one-hour delivery as a competitive moat. It isn't a moat. It’s a suicide pact.
The logistics cost of moving a $12 bottle of detergent from a warehouse to a front door in 60 minutes defies the laws of physics and economics. To achieve this, Amazon (and anyone foolish enough to chase them) has to decentralize inventory to an absurd degree.
- Inventory Bloat: You can’t have one-hour delivery if the item is in a regional hub. It has to be in a "micro-fulfillment center" in a high-rent urban district.
- Labor Fragility: The "last mile" is the most expensive part of the chain. Shrinking the window to an hour removes all ability to optimize routes. You aren't delivering 50 packages on a loop; you’re sending a driver out for one specific panic-buy.
- The Customer Entitlement Trap: You are training customers to value speed over product quality. This creates a race to the bottom where the retailer with the fastest van wins, even if they lose money on every single transaction.
I’ve looked at the unit economics of rapid delivery startups. Most of them burned through billions because they ignored the "density" problem. Amazon can hide these losses in their AWS margins, but for every other retailer, trying to match "one-hour delivery" is a fast track to bankruptcy.
If you want to actually make money in retail, you should be moving toward slower, more efficient delivery cycles that allow for consolidated shipping. The "one-hour" dream is a vanity metric that will eventually be scaled back once the cost of capital stays high.
Geopolitical Posturing and the Cuba Distraction
The chatter about renewed threats toward Cuba or shifting trade stances is often framed as "market volatility." It’s actually a distraction.
Smart money doesn't care about the rhetoric; it cares about the supply chain. When a political figure threatens a neighbor or a trade partner, the real story isn't the headline. It’s the "near-shoring" shift that has already happened.
The idea that a threat to Cuba or a shift in Caribbean policy will move the needle on the US economy is laughable. The real geopolitical risk is the concentration of semiconductor manufacturing in a single geography. If you’re worried about Cuba, you’re looking at the wrong map. You should be looking at the 100-mile stretch of water between China and Taiwan.
If that goes sideways, Nvidia’s $1T projection doesn't just miss; it becomes zero.
The Fallacy of "People Also Ask"
People often ask: "Is Nvidia still a buy at these levels?"
The question is flawed. You’re asking if a momentum play can keep moving. Maybe. But you’re buying at the top of a Capex cycle. You are betting that the tech giants will keep spending money they aren't yet making. That is a gamble, not an investment.
Another common one: "How will one-hour delivery change my shopping habits?"
It won't change your habits; it will change your prices. Expect "convenience fees" to become the norm. The era of "free and fast" is ending because the math simply does not work in an environment where interest rates aren't at 0%.
The Actionable Truth
Stop chasing the "next big thing" that everyone is already shouting about.
If you want to play the AI boom, stop buying the chipmakers and start looking at the companies that own the power. Data centers are hitting a wall not because of a lack of chips, but because they can’t get enough electricity from the grid to run them.
Energy is the real bottleneck. Not GPUs.
And as for retail? The winners won't be the ones who get a package to you in 59 minutes. They will be the ones who figure out how to make you come to the store again, offloading the "last mile" cost back onto the consumer.
The "Morning Squawk" is for people who want to feel informed while they follow the crowd off a cliff. If you want to stay dry, start looking at where the crowd isn't looking.
The infrastructure is built. The party is over. Now comes the bill.
Stop buying the hype and start looking for the exit.
