Hey Traders,

AI is pulling the market in two directions right now. It's driving a building boom measured in hundreds of billions of dollars, and it's quietly changing how people find things online.

Both of those show up in this week's picks. I'm buying a group the market sold like the construction's about to stop. And I'm walking away from a company that makes its money on a road AI agents are learning to skip.

One to Love: Somebody Has to Pour the Concrete

Friday's jobs number should let some air out of the worry that's been leaking into the bond market. Employers added just 29,000 jobs in September against roughly 90,000 expected, and unemployment ticked up to 4.2 percent. Right after the release, the odds of an October rate hike dropped to 14 percent from about 70 percent earlier in the week.

That matters for the Industrial Select Sector SPDR ETF (XLI).

Yes, rates are sticky. The 10-year Treasury yield still finished Friday higher, around 5.28 percent, a few ticks off its highest level since 2002.

I've said it before and I'll say it again… the AI infrastructure buildout is a massive force that keeps rates up, and it isn't going away. But that's a reason to own the builders, not run from them.

Industrials got thrown out with the bathwater. XLI fell about 4 percent in September and more than 8 percent over the last three months, while tech finished as the only sector in the green for September. This group's oversold (sold so hard and so fast that the price has run way ahead of the news).

The buildout has to continue, one way or another. Government and industry will incentivize it and, frankly, force it. Nobody in power lets this stall.

The money's already on the table. Alphabet, Amazon, Meta, Microsoft and Oracle are on pace to spend about $780 billion on capital projects this year, nearly double 2025, and forecasts call for more than $1 trillion a year starting in 2027. Caterpillar and GE Vernova, which sell the generators and turbines these campuses run on, make up almost 12 percent of XLI by themselves.

I think the gloom hanging over data centers right now is largely manufactured. A lot of the fear around AI gets seeded on purpose, at home and abroad, to sow frustration and slow things down. Some of it has even come straight from leaders in the AI space, which still makes no sense to me.

That rhetoric softens from here, and the sentiment turns.

So build we must. And when you have to build, you buy the people who build. One to love.

One to Leave: Nobody Stops to Look Anymore

Start with the chart on AppLovin (APP). It looks terrible. The stock closed September at $290, a hair above its 52-week low and roughly 60 percent below its 52-week high of $738.

When a chart breaks like that, the sellers usually know something before the story's obvious. Online advertising is heading into real upheaval, and AppLovin sits right in the blast radius.

The agentic wave (AI that goes out and handles the task for you) that I think runs for years is built on one motion… pulling information to people. The agent goes and gets it. You don't scroll or wander ten pages deep anymore.

That's a problem for advertisers, because the wandering is the product they pay for. Fewer eyeballs moving across the open internet means less to sell. And an agent fetching an answer doesn't stop to look at a banner or impulse-install a game. It routes right past the ad.

You can already see it in Google's books. In the second quarter, Search revenue on Google's own pages grew 17 percent, while Google Network (the ads Google places on other people's sites and apps) was the only line in the whole company that shrank. Over the same stretch, data from Ozone showed U.S. publisher ad requests down roughly a third from a year earlier.

AppLovin's whole business is being the machine that places and optimizes ads inside other people's apps and games. It's the middleman. It doesn't own the destination or the audience. It rents attention that lives somewhere else and sells access to it.

Compare that to Meta. Meta's a leader in the space, sure… but the real edge is that Meta owns the platforms and the people on them. It can build its own agentic funnels and push you, or your agent, through its own doors.

It's already walking into AppLovin's backyard. Last month Meta rolled out early-access AI tools that turn a text prompt into a playable mobile game, and mobile games are where AppLovin makes its living. If the open ad web shrinks, Meta still owns the room everyone's standing in. That's a fallback.

AppLovin has no room of its own and no owned funnel to retreat into. It's a pure play on exactly the layer agentic flow is built to skip. When the thing you optimize is the thing getting routed around, there's nowhere to hide.

That's why I'm out. One to leave.

So that's where I land this week.

I'm buying the builders in XLI while the selling's overdone, and I'm stepping aside on AppLovin because agents are built to route around the exact attention it rents.

Picking the stock is only part of the job, though.

Every week on the Ticker Highlight Show, I sit down with the rest of the Option Pit crew. And this week we’re bringing on a special guest to spice things up a bit.

That means you’ll get to hear 12 different stock recommendations, and as a member of the Ticker Highlight Show Premium, you get the single best option play out of the dozen.

Tap here to join Ticker Highlight Show Premium.

Here for a good time… AND a long time,

Hans