People keep talking about AMD like it's just "the #2 GPU company," but that's missing the bigger picture.
AMD isn't a one-product company. It has growing businesses in AI accelerators, EPYC server CPUs, client PCs, gaming, and embedded chips. The AI business is still in the early innings, with the MI350 now ramping and the MI400 platform expected to be a major step forward.
The company also has one of the best CEOs in tech. Lisa Su took AMD from a company trading around $2 a share to one competing with Intel in CPUs and Nvidia in AI. Under her leadership, AMD has consistently executed, taken market share, and built a much stronger balance sheet.
What really gives me conviction is that compute demand isn't slowing down. AI, cloud computing, enterprise servers, autonomous systems, and edge computing all require more chips over time. AMD is positioned across multiple parts of that ecosystem instead of relying on a single product.
The market is also acting like Nvidia will win 100% of AI forever. That's not how semiconductors have ever worked. The biggest cloud companies want multiple suppliers for cost, supply chain resilience, and negotiating power. AMD is already supplying products to major hyperscalers like Google and Oracle and continues expanding those relationships.
Could AMD fall 30-50%? Absolutely. Semiconductor stocks are volatile. But volatility isn't the same thing as a broken thesis.
I'm not invested because I think the stock only goes up. I'm invested because I believe AMD will be a much larger and more profitable company five years from now than it is today. If that happens, the stock price will eventually reflect it.
Another thing people ignore is execution. Every product cycle under Lisa Su has made AMD more competitive. Zen transformed CPUs. EPYC took server share from Intel. Ryzen became one of the top consumer CPU brands. Now they're applying that same execution to AI. That's why I trust management.
People also focus way too much on today's earnings instead of future earnings. AMD isn't being valued on what it earned last quarter—it's being valued on what it could earn several years from now if AI accelerator revenue, server CPUs, and software continue scaling. That's why I care more about execution than today's P/E ratio.
The market also acts like Nvidia has to lose for AMD to win. It doesn't. The AI market is becoming so large that multiple companies can grow at the same time. Cloud providers like Microsoft, Google, Oracle, Amazon, and Meta aren't going to rely on a single supplier forever. They want competition, lower costs, and supply chain diversification. AMD is already building relationships with many of those customers.
What really gives me conviction is that compute demand isn't slowing down. AI, cloud computing, enterprise servers, autonomous systems, and edge computing all require more chips over time. AMD is positioned across multiple parts of that ecosystem instead of relying on a single product.