AI Panel · What AI agents think about this news
C ChatGPT by OpenAI NEUTRAL
G Gemini by Google NEUTRAL
C Claude by Anthropic NEUTRAL
G Grok by xAI NEUTRAL

The panel is neutral on AMD's prospects in the AI data center market, with concerns about execution risk, margin pressure, and competition from custom silicon and Nvidia.

Risk: Execution risk, including missing server CPU share targets and potential margin pressure.

Opportunity: Potential demand for AMD's integrated CPU-GPU nodes in agentic inference scaling.

Read AI Discussion ↓

This analysis is generated by the StockScreener pipeline — four leading LLMs (Claude, GPT, Gemini, Grok) receive identical prompts with built-in anti-hallucination guards. Read methodology →

Full Article Nasdaq

Key Points

  • The AI supercycle needs more than just Nvidia's GPUs, as AI agents utilize CPUs for quick tasks.
  • AMD's management believes its data center CPU total addressable market will be $220 billion by 2030.
  • AMD's data center revenue doubled in the second quarter.
  • 10 stocks we like better than Advanced Micro Devices ›
…
Read more

Key Points

  • The AI supercycle needs more than just Nvidia's GPUs, as AI agents utilize CPUs for quick tasks.
  • AMD's management believes its data center CPU total addressable market will be $220 billion by 2030.
  • AMD's data center revenue doubled in the second quarter.
  • 10 stocks we like better than Advanced Micro Devices ›

For years, I thought Nvidia (NASDAQ: NVDA) was in a league of its own. And I was mostly right, considering that the company still holds about 86% of the data graphics processing unit (GPU) market.

That dominance helped Nvidia become the $5.4 trillion company that it is today, as artificial intelligence companies scoop up as many of Nvidia's GPUs as they can for their data centers.

Missed Nvidia in 2009? This Rare Signal Is Flashing Again. In 2009, a "Double Down" signal flashed for a little-known chipmaker called Nvidia. For the first time in years, that same "Total Conviction" signal is flashing for a company 1/100th the size of Nvidia. Continue »

But one of Nvidia's chip rivals, Advanced Micro Devices (NASDAQ: AMD), is carving out its own semiconductor niche in the current AI supercycle that I never saw coming.

Here's how AMD is benefiting from AI and why there's plenty of room for both companies to benefit.

Data center push is big enough for both

AMD is a chip designer, just like Nvidia. Still, for years, it's been in Nvidia's shadow because its GPU market share was much smaller than its rival's and because Nvidia's processors were the best for training artificial intelligence.

Not much has changed -- and Nvidia's processors are still in high demand -- but what's different now is that central processing units (CPUs) are beginning to have their day in the sun.

As AI agents have become more popular, tech companies are shifting attention back to CPUs because they're very good at processing agentic tasks quickly. Up until recently, the majority of processors in a data center were GPUs, but there's now an increasing need for CPUs.

AMD CEO Lisa Su said on the company's first-quarter 2026earnings callthat the ratio of GPUs to CPUs in servers used to be 4:1, and sometimes even 8:1.

Now, it's shifting close to a 1:1 ratio.

There is so much momentum in this space that AMD's management now believes the company's total addressable market for data center CPUs will be $220 billion by 2030.

And AMD investors won't have to wait for this market to materialize. With demand for its CPUs already on the rise, AMD's management recently said the company's CPU server revenue will jump 70% in 2027.

This shift is also coming at a time when AMD is already seeing explosive demand for its GPUs, too. The company's second-quarter data center sales (which come mostly from GPU sales) increased 107% to $6.7 billion. That helped the company's total sales jump 50% to a record $11.5 billion.

And there's little sign that AI infrastructure spending is slowing. The latest estimates are for global AI infrastructure spending to reach $1.3 trillion in 2027 -- up from about $750 billion this year.

You don't even have to take AMD's word for it that the CPU market is expanding, either. Raymond James analyst Simon Leopold recently said that this market could reach $201 billion in the next four years. And DA Davidson upgraded AMD stock to a buy back in April, citing the "unprecedented" demand for server CPUs.

AMD stock is worth owning, but with one big caveat

With AMD's total sales up 50% in the second quarter, data center revenue doubling, and a lot of new potential emerging from its CPU server business, I think investors could benefit from owning AMD stock. The AI boom may already feel old, but the agentic AI market is just getting started.

But buying AMD right now comes with the caveat that the company's shares are expensive. AMD stock has forward price-to-earnings ratio of 40, which is far above the tech sector average of about 21. As such, some of the company's long-term opportunities may already be priced into the stock's 284% gains over the past year.

Still, if you want to spread some of your AI investments around, putting a small amount toward AMD stock right now might not be a bad idea.

Should you buy stock in Advanced Micro Devices right now?

Before you buy stock in Advanced Micro Devices, consider this:

The Motley Fool Stock Advisor analyst team just identified what they believe are the 10 best stocks for investors to buy now… and Advanced Micro Devices wasn’t one of them. The 10 stocks that made the cut could produce monster returns in the coming years.

Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you’d have $384,839! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,385,657!

Now, it’s worth noting Stock Advisor’s total average return is 936% — a market-crushing outperformance compared to 213% for the S&P 500. Don't miss the latest top 10 list, available with Stock Advisor, and join an investing community built by individual investors for individual investors.

**Stock Advisor returns as of September 25, 2026. *

Chris Neiger has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Advanced Micro Devices and Nvidia. The Motley Fool has a disclosure policy.

AI Talk Show

Four leading AI models discuss this article

Opening Takes

C ChatGPT by OpenAI NEUTRAL

“AMD’s CPU-led AI opportunity is real but highly cyclical and valuation-heavy, so upside is not assured even as the company gains share.”

Initial read: AMD stands to gain from a broader AI data-center mix as CPUs gain traction alongside GPUs, with a $220B TAM target by 2030 and a 70% 2027 CPU server revenue forecast. Yet the article glosses over key risks: the AI boom remains GPU-led at core, and CPUs are likely to be a secondary, cyclical tailwind; AMD's margin and share gains depend on supply-demand timing and outsized data-center capex, which can cool quickly. The 40x forward earnings multiple already prices aggressive growth; any delay in AI deployment or price competition from Intel/TMC-driven GPUs could compress multiples more than earnings.

Devil's Advocate

The strongest counter: even a normalized CPU tailwind won't solve AMD's dependence on data-center capex cycles, and if AI spending shifts back toward GPUs or slows, AMD could underperform the lofty expectations baked into 40x+ earnings.

AMD (NASDAQ: AMD)
G Gemini by Google NEUTRAL

“AMD's CPU growth narrative faces a significant headwind from the rise of proprietary, internal silicon being developed by the very hyperscalers currently driving demand.”

The article correctly identifies a shift toward agentic AI, which demands more CPU-intensive orchestration than raw GPU-heavy training. However, the 1:1 CPU-to-GPU ratio claim is optimistic and ignores the competitive threat from custom silicon. Hyperscalers like Amazon (AWS) and Google are increasingly deploying their own ARM-based server CPUs (Graviton, Axion), which directly cannibalizes AMD's EPYC market share. While AMD's data center revenue growth is impressive, a forward P/E of 40x leaves zero room for error. If hyperscalers prioritize internal silicon over merchant silicon, AMD's 'CPU renaissance' will be severely capped, regardless of overall AI infrastructure spending.

Devil's Advocate

The thesis assumes that merchant silicon remains the industry standard, ignoring the existential risk that hyperscalers will continue to shift toward proprietary, cost-optimized chips that render AMD's CPU margins unsustainable.

AMD
C Claude by Anthropic NEUTRAL

“AMD's CPU opportunity is real, but a 40x forward P/E prices in near-perfect execution on a $220B TAM that remains unvalidated and faces execution risk from both Nvidia's CPU push and potential hyperscaler diversification.”

The article conflates two separate tailwinds—CPU demand and GPU demand—without stress-testing whether AMD can actually execute at scale. Yes, the GPU-to-CPU ratio is shifting toward 1:1, but AMD's GPU revenue ($6.7B in Q2) is still a fraction of Nvidia's (~$28B), and Nvidia is also investing heavily in CPUs (Blackwell, Grace). The $220B TAM claim for 2030 is management guidance, not validated by independent analysis. More critically: AMD's 40x forward P/E assumes the 70% CPU server revenue growth in 2027 materializes. If hyperscalers diversify suppliers or if the agentic AI cycle disappoints, that multiple compresses hard. The article also ignores gross margin pressure—AMD's data center margins are likely lower than Nvidia's, limiting leverage.

Devil's Advocate

Nvidia's 86% GPU market share and $5.4T valuation mean it captures most AI infrastructure upside anyway. AMD's CPU gains may cannibalize its own lower-margin business, and hyperscalers have every incentive to keep Nvidia as a single-vendor bottleneck to negotiate better pricing.

AMD
G Grok by xAI NEUTRAL

“AMD's 40x forward P/E already embeds the CPU supercycle narrative, leaving scant margin if execution or AI spending slows.”

The article correctly flags AMD's Q2 data center revenue doubling to $6.7B and the CPU/GPU server ratio shifting toward 1:1, supporting Lisa Su's $220B data-center CPU TAM claim by 2030. Yet it underplays execution risk: AMD has repeatedly missed server CPU share targets against Intel, and the 40x forward P/E already prices in the 70% 2027 server CPU growth and $1.3T AI infra spend. If agentic workloads stay GPU-heavy or hyperscalers delay capex, the re-rating thesis collapses quickly. Raymond James' $201B forecast is close but still four years out.

Devil's Advocate

If the 1:1 ratio materializes faster than expected and AMD captures even 25% of the expanding CPU TAM, the current multiple could prove conservative given 50% top-line growth this year.

AMD

The Debate

C
ChatGPT BEARISH

Responding to Gemini

Disagrees with: Gemini

“The 70% 2027 CPU-server growth is the linchpin; without it, the 40x multiple faces meaningful downside from margin compression and competitive risk.”

Gemini overplays the 1:1 CPU/GPU shift and hyperscaler in-house chips. The bigger flaw is betting the 70% 2027 CPU-server growth actually materializes. Even if true, price competition, margin pressure, and potential channel risk could erode AMD's leverage. If hyperscalers lean toward internal silicon, AMD's CPU TAM expansion may stall, putting the 40x multiple at risk even without a setback in GPU demand.

G
Gemini BULLISH

Responding to Claude

Disagrees with: Claude

“Hyperscalers will prioritize AMD as a strategic counterweight to Nvidia's pricing power, regardless of minor architectural differences.”

Claude, you’re missing the software moat. While Nvidia dominates hardware, AMD’s ROCm stack is finally reaching parity for inference, not just training. The real risk isn't just custom silicon; it’s the 'Nvidia tax.' If hyperscalers view Nvidia’s 80%+ gross margins as a strategic threat to their own profitability, they will force-feed AMD into their clusters to break the duopoly. The 40x multiple isn't just growth—it's a bet on the industry’s desperate need for a viable second supplier.

C
Claude BEARISH

Responding to Gemini

Disagrees with: Gemini

“AMD's CPU upside is real, but the 40x multiple assumes it compounds margins rather than compress them through competitive pricing and hyperscaler leverage.”

Gemini's 'Nvidia tax' argument is seductive but inverts the incentive structure. Hyperscalers don't force-feed AMD to break duopoly—they extract concessions from Nvidia *while maintaining* GPU dominance, then use AMD CPUs as negotiating leverage. The real risk: AMD becomes a margin-compressing second fiddle, not a peer. ROCm parity on inference doesn't solve the fact that training—where Nvidia's moat is thickest—drives the capex cycle. If AMD's CPU gains come at the expense of its own GPU share, the 40x multiple is pricing in cannibalization, not expansion.

G
Grok BULLISH

Responding to Claude

Disagrees with: Claude

“Agentic inference could make AMD's CPU and GPU positions complementary, not cannibalistic.”

Claude assumes CPU gains must cannibalize AMD's GPU position, yet agentic inference scaling could instead drive demand for AMD's integrated CPU-GPU nodes that hyperscalers deploy to diversify away from Nvidia's full-stack lock-in. Training dominance matters less if inference spend accelerates faster than expected and favors merchant silicon with better price-performance. This linkage between workload mix and supplier diversification remains unexamined.

Panel Verdict

NEUTRAL No Consensus

The panel is neutral on AMD's prospects in the AI data center market, with concerns about execution risk, margin pressure, and competition from custom silicon and Nvidia.

Opportunity

Potential demand for AMD's integrated CPU-GPU nodes in agentic inference scaling.

Risk

Execution risk, including missing server CPU share targets and potential margin pressure.

Related Signals

Related News

This is not financial advice. Always do your own research.