Hello there.

Happy Friday! Hear about that huge IPO party this year? Oh, you didn’t get invited? Yeah, me neither.

In this issue:

We don’t need $3T to have a good time; let’s throw our own party. We’ll bring the dip (I mean, buy the dip).

In partnership with MedX

Please support our partners!

Wall Street has seen big IPOs before. Facebook, Alibaba, you know the types. This year, it looks like we're getting OpenAI, SpaceX (merged with xAI), and Anthropic hitting public markets, potentially in the same window, and carrying a combined valuation somewhere around $2.9 trillion. To put that in perspective, the entire US IPO market raised $469 billion from 2016 to 2025. These three companies alone could demand half that in a single quarter.

Here's where it gets messy. When companies go public, they usually sell around 15-25% of themselves to everyday investors. But these companies are SO massive that doing that would mean raising like half a trillion dollars all at once. But that's just not possible. So instead, they'll probably only sell a tiny slice, maybe 3-8%.

The problem is, to get added to the S&P 500, you need to sell at least 50% to the public. So these companies won't qualify right away. Or at least, they wouldn’t… unless the S&P breaks its own rules?

But regardless of how, what happens when they eventually DO get in? Every single index fund in America has to buy them. And those funds don't just have cash sitting around. They have to sell OTHER big stocks to make room. Imagine SpaceX showing up as one of the biggest companies in the entire index. The scramble to buy it could shake up literally everything else.

We’re looking at FOMO season for the everyman. Pre-IPO investment vehicles are popping up everywhere, and demand for early access is pushing valuations into nosebleed territory before the tickers even exist. But keep in mind: buying on IPO day rarely captures the real gains. The founders and VCs who got in early are about to print generational wealth. Everyone else is fighting for scraps after the opening bell. History says the post-IPO hangover can be brutal, but history also never saw three trillion-dollar AI companies hitting the market at once. This is uncharted territory, and the ripple effects will be felt for years. Stay safe out there.

Building AI requires more than code. It requires data centers. And data centers require land, power, and thousands of workers willing to relocate to the middle of nowhere. Enter the "man camp," a relic from the shale oil boom now getting a Silicon Valley rebrand as developers budget $700 billion for data center expansion in 2026 alone.

Picture this: a 1.6-gigawatt data center rising in rural Dickens County, Texas. The workers building it live on-site in gray housing units complete with gyms, game rooms, laundromats, and cafeterias, grilling ribeyes to order.

Target Hospitality's chief commercial officer describes the AI construction pipeline as "the largest, most actionable" opportunity he's ever seen. At $132 million just for the Dickens County camp, that math tracks. But the model raises some… problematic memories. Company towns didn't die because they were inefficient. They died because workers eventually realized that your employer owning your housing, food, and recreation creates a power imbalance that tends to explode. The Pullman Strike of 1894 happened for a reason. Whether "workplace hubs" in the desert turn into utopia or tinderbox depends entirely on oversight that, historically, tends to arrive after something goes very wrong.

In this episode of The Best One Yet, you’ll get the lowdown on Jensen Huang’s first blog post in six years and his framework for understanding the entire AI economy. Meanwhile, Lego proves that the best growth strategy might be selling toys to…adults.

📻 Tune in to:

🎧 Listen on:

That’s all for today. Write us and let us know your thoughts on the market, the newsletter, or the weather—we’d just love to hear from you.

Till next time,
— Brandon and Blake