Latest / Stock Picking with Fexingo: Individual Equities, Research, and Building a Concentrated Portfolio / How NVIDIA Ticker NVDA Is Becoming the Oil Trade of AI
Transcript
- Lucas: If you glance at the Nasdaq this week — down one and a half percent over five days — you'd be forgiven for thinking AI is cooling off. Meta off six percent, Alphabet off eight, Tesla off sixteen. But then you see NVIDIA, ticker N-V-D-A, up nearly three percent on the week. And that divergence is worth a long look. Luna: It is striking. The poster child of the AI trade is actually gaining while almost everything else in tech is sliding. What's driving that? Lucas: Part of it is the oil story. You've got crude surging toward triple digits, and suddenly energy stocks are the place to be — ExxonMobil up six and a half percent this week alone. But here's the thing: NVIDIA is increasingly being treated by the market as an energy play, not just a chip company. Luna: How so? I mean, they make chips that consume power, they don't produce oil. Lucas: Right, but the thesis is that AI data centers are becoming massive consumers of electricity. Goldman Sachs put out a note in June estimating that data center power demand will grow at a compound annual growth rate of fifteen percent through 2030. And NVIDIA's H100 and next-gen Blackwell chips are the engines inside those data centers. So as energy prices rise, the cost of running AI goes up — but NVIDIA's pricing power goes up too, because their chips are the most efficient per watt for training large models. Luna: So they're almost a hedge on the energy theme. You own Exxon for the supply side, you own NVIDIA for the demand side. Lucas: Exactly. And there's another leg to this that I think is underappreciated. Sovereign AI — governments building their own national AI infrastructure. NVIDIA has been signing deals with countries like India, Japan, and the UAE to supply entire AI supercomputers. These are multi-billion dollar contracts that aren't tied to the quarterly whims of Silicon Valley. They're tied to national security budgets. Luna: That's a different demand profile. It's more like defense spending than consumer tech. Lucas: Exactly. And that makes NVIDIA's revenue stream more resilient than, say, Alphabet's ad revenue or Meta's ad revenue when the economy softens. Now, let's talk about the valuation because I know that's the pushback everyone gives: 'NVIDIA at a price to earnings ratio of fifty is expensive.' Luna: It is high. But if earnings keep growing at fifty percent year-over-year, that P/E compresses pretty fast. Lucas: That's the bull case in a nutshell. But I want to focus on one specific number that I think is more telling than the P/E. NVIDIA's free cash flow yield. Right now, even at this stock price, they're generating free cash flow — after all capital expenditures — of about three and a half percent. That's not far from the ten-year Treasury yield. And their free cash flow is growing. Luna: So you're saying the stock is priced as if it's a bond with a growth kicker? Lucas: In a way, yes. And look at what the company is doing with that cash. They've ramped up buybacks — bought back over eighteen billion dollars worth of stock in the first half of this year. That's a signal that management thinks the stock is undervalued even at these levels. Luna: But there's a risk too, right? If AI spending slows, or if a competitor like AMD or a custom chip from a hyperscaler eats into their market share, that cash flow narrative breaks. Lucas: Absolutely. And I think the market is pricing in that risk, which is why the stock hasn't gone parabolic again. But here's what I find interesting: the downside scenarios people paint — like 'AI is a bubble' or 'NVIDIA's gross margins will compress' — those have been the bear case for eighteen months now. And every quarter, NVIDIA has beaten expectations. The burden of proof is shifting to the bears. Luna: So if you're building a concentrated portfolio, how do you think about NVIDIA alongside something like Exxon? They're in totally different sectors. Lucas: I think they're complementary. Exxon gives you exposure to the physical energy supply chain — oil and gas that the world still needs. NVIDIA gives you exposure to the digitization of energy demand. If you believe that AI is a structural shift, not a cyclical fad, then owning both makes sense. It's not either-or. It's both. Luna: And the market seems to agree this week. Energy and select AI infrastructure are the pockets of strength while everything else wobbles. Lucas: Precisely. Now, there's a third leg I want to touch on quickly — industrial robotics. NVIDIA's Jetson platform for edge AI is finding its way into factory automation. They recently announced a partnership with an industrial robotics firm — I won't name them because they're not public — but the thesis is that manufacturing is going to be a massive AI adoption driver as companies reshore supply chains. And that's a long-tail opportunity that doesn't depend on ChatGPT usage numbers. Luna: That's a good point. The narrative around AI has been so consumer- and cloud-focused that we forget about the physical world. Lucas: Right. And that physical world AI market is still very early. So if today's conversation was useful to you — if it changed how you think about positioning for that shift — I'll just mention that the way we keep these episodes ad-free is through listener support. If you're finding value here, you can buy us a coffee at buy me a coffee dot com slash fexingo. It's a small gesture that keeps the lights on and lets us keep digging into these specific angles. Luna: Yeah, and honestly, we get emails from listeners saying they used a thesis from the show to make a trade or rethink a portfolio. That's the kind of impact we're aiming for. So if you're one of those people, thank you. And if you want to help more people find it, that coffee link is the way. Lucas: Alright, back to NVIDIA. One last number I want to leave you with. NVIDIA's revenue from its data center segment in the most recent quarter was twenty-six billion dollars. That's up one hundred and forty percent year-over-year. And yet the stock is roughly flat from its peak in June. So the market is saying, 'We see it, but we need to see more.' The next catalyst could be the Blackwell ramp in the second half of this year. Luna: That's a big number. And if Blackwell drives another leg of growth, the stock could re-rate higher. Lucas: It could. But as always, it's about the execution. For now, NVIDIA is showing that it's more than just a hype stock — it's becoming an infrastructure bet on two of the biggest themes in the market: energy and AI. And in a week where the Nasdaq is down, that's worth paying attention to.