AMD’s AI Bet: Why Stock Valuations Don’t Always Match Reality

AMD just released news about their AI infrastructure capabilities, and naturally, the stock jumped. Then the skeptics showed up. According to some analysts, AMD might be overvalued by as much as 38% based on these announcements. As someone who’s built products and watched markets, I find this tension genuinely interesting—not because the number matters, but because it reveals something fundamental about how we value tech companies.

Let me be clear: I’m not here to tell you whether AMD’s stock will go up or down. I’m no fortune teller. But I can talk about what I see happening in the real world with AI infrastructure, and why the gap between hype and actual value is worth understanding.

The AI Infrastructure Gold Rush Is Real

Here’s what’s actually happening right now. Every company building AI products needs chips. Lots of them. For the past couple of years, NVIDIA has dominated this market almost completely. They’ve been the only realistic choice for serious AI workloads. But that’s starting to change.

AMD has been quietly building competitive AI chips. Their EPYC processors and custom AI accelerators are getting better. More importantly, they’re actually becoming viable alternatives. I’ve talked to engineers at multiple startups who are now seriously evaluating AMD hardware instead of automatically going with NVIDIA.

This is huge. It means AMD isn’t just dreaming about market share—they’re actually taking it. Or at least they have a legitimate shot at taking it. That’s not nothing.

The problem isn’t whether AI infrastructure demand exists. It does. The problem is figuring out how much of that demand actually flows to AMD versus staying with NVIDIA or going elsewhere.

Where Valuations Get Disconnected From Reality

Stock price movements often overreact to news. Especially AI news. We’ve all seen it. A company mentions AI, and investors start pricing in scenarios where AI drives 50% revenue growth within three years. Sometimes that happens. Usually it doesn’t.

The 38% overvaluation claim essentially says: “The market is pricing in too much future success from AMD’s AI play.” Fair point worth considering. But here’s what actually matters when you’re trying to build a real business:

• Can AMD actually manufacture these chips at scale? (Yes, they have fabs and partners.)
• Are customers willing to switch from NVIDIA? (Some are. Not all.)
• Can AMD support the software ecosystem? (Getting better, but still behind.)
• Will margins hold up? (That’s the real question.)

The valuation question assumes the market overestimates AMD’s ability to execute on points 2, 3, and 4. Maybe they’re right. Maybe not. What I know from building products is that execution is messy. Things take longer than expected. But sometimes they work out better than expected too.

The Long View vs. The Stock Price View

Here’s my take, based on watching dozens of companies try to disrupt entrenched players: AMD isn’t trying to steal 100% of NVIDIA’s market. They don’t need to. They just need to own a significant chunk of a massive market that’s only growing.

Whether AMD’s stock is fairly valued today matters less than whether their business is actually improving. And from what I see, it is. They’re shipping real products. Customers are buying them. Engineers are testing them seriously instead of dismissing them immediately.

Could the stock pull back 38%? Sure. Stock prices move all the time for all kinds of reasons. But that’s different from whether AMD’s core business is sound.

If you’re an investor trying to time AMD’s stock price, you need to make your own bet on execution and market timing. If you’re interested in the actual tech, the actual competition, and how the AI chip market evolves—that’s where the real story is. And that story is still being written.

AMD’s making real moves. Whether the stock is fairly priced is somebody else’s problem.

Source: AMD (AMD) Stock Could Be 38% Overvalued On Fresh AI Infrastructure News