THE SECOND FAVORITE

(NVDA), (AMD), (META), (MSFT), (ORCL), (MRVL), (MDTTF), (QCOM), (INTC)

Everyone’s chasing Nvidia (NVDA) like it’s the last helicopter out of Saigon. Meanwhile, a quieter story is unfolding — one we’d be wise to pay attention to. 

While the crowd piles into Nvidia like it’s the only game in town, a few sharper players are quietly slipping Advanced Micro Devices (AMD) chips onto the table.

I learned this lesson the hard way back in 2018 when my buddy Dave went all-in on a moonshot play. Six months later, he was on a strict ramen diet, while I was enjoying meals that required actual utensils. 

Lesson? Even in a booming market, diversification isn’t optional. It’s oxygen.

Fast-forward to today’s AI arms race: Nvidia has the biggest guns — 85% market share and a developer base that borders on cultish. 

But AMD is no longer just catching up. It’s shaping up to be the most compelling underdog in tech. It’s the sleeper hit the critics love. Meta (META), Microsoft (MSFT), and Oracle (ORCL) are all starting to nibble.

Why? AMD’s MI355X chips already deliver 40% more tokens per dollar than Nvidia’s. In fact, Sam Altman from OpenAI called the MI450s “totally crazy.” When the guy building the future starts cheering for the underdog, it’s worth noting.

Here’s where it gets interesting: AI compute needs are diversifying. 

Training, inference, and edge computing all demand unique hardware setups, power profiles, and pricing strategies. 

Nvidia still dominates training, but inference — aka the workhorse of real-world AI — is still up for grabs. AMD’s edge in price-performance makes it a strong contender here.

Even a modest share of inference workloads could mean billions in new revenue for AMD. And its open-source ROCm software stack? It’s gaining traction, quietly eroding CUDA’s once-impregnable moat.

Now, this isn’t a David-versus-Goliath fantasy. 

AMD doesn’t need to topple Nvidia to win. With the AI chip market racing toward $500 billion, just catching the overflow from Nvidia’s supply constraints is a major opportunity.

And AMD is uniquely positioned to do just that. Its open standards make it easier for companies to integrate without being locked into proprietary ecosystems. 

Nvidia’s CUDA is powerful, yes, but also exclusive — a walled garden. AMD’s openness is attracting heavyweight partners like Marvell (MRVL), MediaTek (MDTTF), and Qualcomm (QCOM). Even Microsoft is co-developing chips with them. 

To be clear: Nvidia still holds the throne. Its market share and developer loyalty are formidable. 

But AMD isn’t trying to stage a coup. It’s carving out its own path. This is more of an evolution than a revolution.

And the market’s starting to notice. AMD shares are down 18.6% this year, and its forward P/E has compressed 30%. Yet, fundamentals are strengthening. 

Translation: this might be the market handing you a gift.

Lisa Su has led a turnaround worthy of an HBS case study. From near-collapse to credible Nvidia rival, she’s engineered a strategy that’s winning traction. 

Oracle (ORCL) has deployed over 131,000 MI355X chips. Meta and Microsoft are already integrating AMD into their AI stacks. This isn’t buzz. It’s real adoption.

And adoption matters. 

In tech, relevance begets more relevance. Once a chip provider gets baked into large-scale AI workflows, momentum builds. 

More usage leads to more developer support, which feeds back into confidence and capability. It’s the network effect in silicon.

From a portfolio perspective, AMD is what I call an “asymmetric opportunity”: limited downside, meaningful upside. 

If you’re building a smart AI portfolio, this is how you hedge intelligently. No need to abandon Nvidia. But concentrating entirely in one name? That’s a rookie move.

Plus, AMD’s strength isn’t limited to AI. Its client CPU business jumped 68% year-over-year, clawing share away from Intel (INTC). 

So what you’re really buying is exposure to both the AI megatrend and a broader semiconductor recovery. That’s two shots on goal with one investment.

Technically, AMD is setting up for a breakout. With RSI at 67.9, it’s building momentum but still has headroom before hitting overbought territory. The chart setup resembles AMD’s 2022 rally, and we know how that ended.

Bottom line: AMD at $126–130 isn’t a speculative moonshot. It’s a well-positioned play on the changing dynamics of AI hardware. 

Will it dethrone Nvidia? Probably not. Does it need to? Absolutely not. 

If it climbs to $200+ by picking up market share while Nvidia stays dominant, that’s a win in any portfolio.

And let’s be honest: ramen’s cheap, but I prefer steak. So hedge accordingly.