Pence: NVDA "Pivotal Chokepoint" of AI & Likes MSFT, AMZN, GOOGL in Mag 7 artwork

Pence: NVDA "Pivotal Chokepoint" of AI & Likes MSFT, AMZN, GOOGL in Mag 7

Schwab Network

August 18, 2026

Dryden Pence returns to the Watch List and walks investors through some of his top stock picks, which includes Nvidia (NVDA), a company he calls a "pivotal chokepoint" in AI. He doesn't see the AI trade slowing down and sees Nvidia capitalizing on every aspect of the industry.
Speakers: Dryden Pence

Topics: Investing, Business

**SPEAKER_1** (0:00)
Thank you so much for being with us. So since you have Nvidia as a name you like, let's talk about, I mean, they have a $350 price target. Do you see a lot of upside potential for an Nvidia, and if so, why?

**Dryden Pence** (0:11)
I do. I mean, Nvidia still is the absolute choke point in AI, and they're absolutely essential to this growth, which is still quite phenomenal. And when you think about it, in most all these tech companies, what I'm liking is that the E is growing faster than the P. If you look at P-E ratios, earnings are growing faster than prices are. Nvidia is like, what, 22? Take a look at a company that's growing as fast as they are, with the earnings growing as fast as they are, and having a forward P-E of 22 So I think I would agree with Bank of America on this.

**SPEAKER_1** (0:42)
Yeah. And Bank of America actually said that this call here is because of the $105 billion investment commitment with open AI. That's one of the reasons, along with many others, just the shift in the commitment that Nvidia is showing. The benefits of AI, we're waiting for the monetization, but this just shows the trend overall. Is that right?

**Dryden Pence** (1:04)
I think so. But when we talk about AI, you've got to remember, you know, 56 percent of the companies in America now have an AI account, and they're paying for this stuff.
And of those, 71 percent of them are showing increases in labor productivity. So when you think about, and then like maybe only 7 percent of them say that they fully integrate it. So you see this huge opportunity for AI integration, and we're still just building out this infrastructure. And Nvidia and many of these companies are absolutely essential to that. So I think you've got a very strong continued push in what's going on. Because in the end, if AI is increasing labor productivity, people are going to buy it. And if it didn't increase labor productivity, it would be just an interesting, you know, parlor game or in Shea.

**SPEAKER_1** (1:50)
Do you feel like they're being so, I mean, this is not a Bank of America segment, but the price target, we could discuss that. Where you really, 50% upside, really. When they call 350, it's a 219 or so. It certainly shows an aggressive move. I mean, they're not saying 250 Do you really think that some of these tech names can move 20, 30, 40, 50 percent? You have a slew of names that we're going to be discussing in here. I mean, you have Nvidia and Microsoft, two of the sort of the big names that people know and like. Are those the kinds of names that can move 20, 30, 40, 50 percent?

**Dryden Pence** (2:23)
When earnings are moving 20, 30, 40, 50 percent, which is happening, then the price of the stock is record. Record earnings are going to be record stock price.

**SPEAKER_1** (2:32)
You had another one you had in there was LMT, another big name in the big names we know. Then I want to get to some of the details on the equipment. But to mention that one too.

**Dryden Pence** (2:43)
Sure. When we talk about that one, you have to recognize the world is not going to sing Kumbaya. We know that. Wars are here, they're always going to be here. Lockheed Martin has a monopoly on replacing all the Patriot missiles, replacing all the THAAD systems and the F-35.
So they have $230 billion worth of order backlog. That's three years worth of revenue.
Show me a company that's got three years worth of revenue locked in with a government contract. That's probably one you want to buy.

**SPEAKER_1** (3:14)
Understood. Okay. So we covered some of the names that you're watching in the MAG-7.
Let's get to semiconductor equipment and some of the other names that you like in the semi-spotlight. Tell me about some of those names and why this AI build out because you have names that go beyond just the MAG-7.

**Dryden Pence** (3:32)
Exactly.
It's not just the MAG-7, it's all the essential components of the MAG-7. So before we want to talk about it, I call it Test, Test, Connect and See.
What I mean by that is we took the Keysight Technology, which they're up about 66 percent this year. They've got some volatility. They're going to be announcing earnings and stuff. But when you get 28 percent revenue growth, you've got adjusted earnings growth, 42 percent. But AI has to work. And for the chips to work, they have to be tested. You don't want to build one of these data centers and all of a sudden have one or two faulty chips, blow it down. So Keysight really works the testing of that to make sure when the chip goes in, it goes in right. And that's important. Or then when you look at Teradyne, the same thing. Teradyne is a big tester of those things. And again, they're testing to make sure, and they're up 106 percent. I mean, it's an amazing company. And again, it's test equipment. So to make sure it works before it ships. And then you have Lumentum, which are LITE, they're up like 130 percent. And when you think about the technology of moving information between chips, factories, stuff like that, all of this is going over fiber optics really fast. And the technology that does that becomes essential. And then the last one that we've talked about, and while we can circle back to any of these if you want to go deeper, is Teradyne.

3 more minutes of transcript below

Thousands of transcripts fetched by people building searchable podcast archives

Feed this to your agent

Try it now — copy, paste, done:

curl -H "x-api-key: pt_demo" \
  https://spoken.md/transcripts/1000651996090

Works with Claude, ChatGPT, Cursor, and any agent that makes HTTP calls.

From $0.10 per transcript. No subscription. Credits never expire. Prices exclude VAT, added at checkout for EU customers. Not what you expected? Email us within 14 days with 20 or fewer credits used and we refund the pack in full.

Using your own key:

curl -H "x-api-key: YOUR_KEY" \
  https://spoken.md/transcripts/YOUR_EPISODE_ID