If you’ve invested in semiconductor stocks recently, you’re likely celebrating Advanced Micro Devices’ (NASDAQ:AMD) stunning run in 2025. AMD stock has soared 93% year-to-date, handily outpacing the already impressive 36% gain of its longtime rival, Nvidia (NASDAQ:NVDA). For years, Nvidia stock has been the undisputed champion of investor returns. But this breakout performance has many wondering: could this be the year the underdog finally starts to consistently outperform the Goliath of artificial intelligence (AI) computing?
The battle is especially fascinating because it pits two cousins against each other: Nvidia’s visionary founder and CEO, Jensen Huang, against the transformative leader of AMD, Dr. Lisa Su. The prize? Dominance in the explosively growing artificial intelligence infrastructure market.
AMD vs Nvidia: The David vs. Goliath dynamic
Nvidia is a behemoth with a market valuation of US$4.5 trillion (C$6.3 trillion) and annual revenue of US$165 billion (C$231.7 billion). In comparison, AMD, at a US$378 billion (C$530.9 billion) market cap and US$29.6 billion (C$41.6 billion) in revenue, is a mere fraction of its size. This size difference is AMD’s secret weapon.
Here’s why: the law of large numbers makes it incredibly difficult for a giant to double in size. For Nvidia to double its market cap, it must essentially create another company its own size by expanding the entire AI market. AMD, however, only needs to chip away a small portion of Nvidia’s AI accelerator market share to achieve massive growth.
A US$25 billion revenue opportunity, for instance, would represent an 84% surge for AMD but only a 25% bump for Nvidia. This is why AMD’s smaller base is a powerful advantage in the race for growth.
AMD’s game plan: Chipping away at Nvidia’s AI monopoly
AMD is actively executing a plan to win market share from Nvidia, with advanced MI450 AI accelerators hitting the “mass” market in 2026. The company’s recent landmark deal with ChatGPT developer OpenAI is a massive vote of confidence. While specifics are complex, the agreement involves building six gigawatts of AI infrastructure and includes ambitious milestones tied to AMD stock, suggesting OpenAI believes significant upside is achievable.
A single gigawatt AI data center means a US$40-to-US$50-billion revenue opportunity for Nvidia. Even if AMD discounts its offerings, the opportunity is huge. Combined with a recent Oracle (NYSE:ORCL) multi-year deal targeting initial deployment of 50,000 AMD GPUs starting in 2026, the growth opportunity is impressive, more so given AMD’s low revenue base.
Furthermore, AMD has built a complete toolkit to compete with Nvidia’s end-to-end offerings. Its acquisition of ZT Systems enhanced its ability to deliver full-scale AI data center solutions. More critically, AMD is addressing Nvidia’s most formidable moat: its CUDA software ecosystem. CUDA is a programming platform that has locked developers into Nvidia’s hardware for years. AMD’s answer is ROCm, an open software platform that allows customers to run AI models on AMD systems right out of the box. The company claims its GPUs are already being used by seven of the ten largest AI model builders today. The goal is to make switching from Nvidia not just possible, but practical and cost-effective.
Valuation
With a forward price-to-earnings (P/E) ratio of 102, AMD stock is significantly more expensive than Nvidia, which trades at a forward P/E of 41.7. AMD’s recent rally, fueled by speculative optimism around its future deals, may have taken its share price far too high.
However, a more telling metric might be the Price-Earnings-to-Growth (PEG) ratio, which factors in a company’s growth trajectory. Nvidia’s PEG ratio of 0.80 suggests it’s still undervalued given its explosive earnings growth potential. AMD’s PEG of 1.5 indicates it might be fully valued or even overvalued after its year-to-date rally.
Investor takeaway
AMD’s story is incredibly exciting due to recent strategic deals, otherwise the smaller AI stock’s performance was similar to Nvidia’s, until its October press releases ignited a good run.
The AI race is far from over. For AMD to truly outperform Nvidia stock over the next two to three years, it must flawlessly convert its promising deals into massive revenue streams and continue to erode Nvidia’s first-mover advantage. For long-term oriented investors, the choice is between the proven, yet still growing, titan and the agile, ambitious challenger betting everything on a breakthrough. The semiconductor showdown has never been more thrilling.
