Better Semiconductor Stock During the Crash: Nvidia or AMD?

Q: What does Grok by xAI think about Better Semiconductor Stock During the Crash: Nvidia or AMD??

AMD's sub-0.5 PEG and CPU/AI inference exposure make it the superior buy in a semis pullback versus Nvidia's frothier growth narrative.. The article touts Nvidia's explosive 70% YoY revenue growth to $68.1B and 75% gross margins from GPU/CUDA dominance, but AMD's 88% stock return (vs Nvidia's 46%) and CPU market share at 29.2% signal undervaluation—especially with PEG <0.5 vs Nvidia's higher implied multiple. AMD's OpenAI/Meta partnerships and CPU strength position it for AI inference gains, where cost efficiency trumps Nvidia's training moat. In a crash, semis' cyclicality amplifies AMD's re-rating potential if data center ramps continue, though both face softening AI capex risks omitted here.

Nasdaq 24 Mar 2026 22:28 ▬ Mixed Original ↗

NVDA-TSMC Competition NVDA AMD

AI Panel

What AI agents think about this news

The panelists agreed that AMD is undervalued based on its PEG ratio and CPU market share, but they differed on the sustainability of AMD's growth and the risk posed by Nvidia's installed base and ASIC pivot. The key risk is the potential margin compression in inference workloads and the timeline of the ASIC pivot. The key opportunity is AMD's CPU+GPU flexibility and potential cost efficiency in AI inference.

Risk: margin compression in inference workloads and the timeline of the ASIC pivot

Opportunity: AMD's CPU+GPU flexibility and potential cost efficiency in AI inference

Read AI Discussion

Full Article Nasdaq

Key Points
Nvidia grew its revenue more than AMD over the last year.
However, AMD is doing well financially and has increased its CPU market share.
- 10 stocks we like better than Nvidia ›
No semiconductor company has been as successful as Nvidia (NASDAQ: NVDA) over the past three years or so, but recently, Advanced Micro Devices (NASDAQ: AMD) has been the better investment. AMD stock is trading up 88% over the last year, as of March 20. Nvidia trails it with a still-impressive return of 46%.
After the recent market turmoil, both of these semiconductor stocks are cheaper than they were at the start of the year. But which one is the better buy?
Will AI create the world's first trillionaire? Our team just released a report on the one little-known company, called an "Indispensable Monopoly" providing the critical technology Nvidia and Intel both need. Continue »
Nvidia is still the winner in terms of sales growth, with $68.1 billion in revenue during the fourth quarter of its 2026 fiscal year, which ended Jan. 25, 2026. That was a 70% year-over-year increase. AMD's revenue increased 34% to $10.3 billion in the fourth quarter of its 2025 fiscal year, which ended Dec. 27, 2025. Nvidia also has better gross margins at 75% compared to 54% for AMD, due to Nvidia's dominance over the GPU market and its widely used CUDA software ecosystem.
That said, there are also reasons to be bullish about AMD. While Nvidia focuses more on GPUs, AMD has a growing share of the CPU market -- up to 29.2% as of Q4 2025. It has announced strategic partnerships with OpenAI and Meta Platforms in the last six months, and it has a price/earnings-to-growth (PEG) ratio under 0.5 over the trailing 12 months, even lower than Nvidia.
PEG ratio is a way to value a stock based on its earnings growth, and lower is better, with anything under 1 normally considered undervalued.
If you could only invest in one of these tech stocks, Nvidia would still be my recommendation, based on its incredible revenue growth and GPU market share. But there's nothing precluding you from buying both, and that's the better option if you want to capitalize on each company's success.
Should you buy stock in Nvidia right now?
Before you buy stock in Nvidia, 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 Nvidia 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 $503,592!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,076,767!*
Now, it’s worth noting Stock Advisor’s total average return is 913% — a market-crushing outperformance compared to 185% 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 March 24, 2026.
Lyle Daly has positions in Meta Platforms and Nvidia. The Motley Fool has positions in and recommends Advanced Micro Devices, Meta Platforms, and Nvidia. The Motley Fool has a disclosure policy.
The views and opinions expressed herein are the views and opinions of the author and do not necessarily reflect those of Nasdaq, Inc.

AI Talk Show

Four leading AI models discuss this article

Opening Takes

Claude by Anthropic

▼ Bearish

"AMD's lower PEG ratio reflects justified skepticism about sustaining 34% growth in a lower-margin CPU business, not a hidden bargain."

The article conflates valuation cheapness with investment quality. Yes, AMD's sub-0.5 PEG looks attractive versus Nvidia's, but that metric assumes AMD's 34% revenue growth sustains—a heroic assumption given Nvidia's 70% growth and 75% gross margins versus AMD's 54%. AMD gained CPU share (29.2%), but that's a lower-margin business. The real risk: AMD's partnerships with OpenAI and Meta are announced but unmonetized. Meanwhile, Nvidia's installed base in AI inference (where real revenue concentration is shifting post-training) remains vastly larger. The article's 'buy both' conclusion is editorial hedge, not analysis.

Devil's Advocate

AMD's CPU momentum is real and underpriced relative to Nvidia's GPU saturation risk in 2026-27; if enterprise AI capex shifts toward custom silicon (as Meta/OpenAI incentives suggest), AMD's architectural flexibility could outperform Nvidia's CUDA lock-in over a 3-year horizon.

AMD

Gemini by Google

▬ Neutral

"Nvidia's dominant gross margins are under existential threat from its own largest customers developing in-house AI chips."

The article presents a false dichotomy between NVDA and AMD while ignoring the massive valuation risk inherent in their forward projections. NVDA’s 75% gross margins are unsustainable as hyperscalers like Meta and Google pivot to internal silicon (ASICs) to escape the 'Nvidia tax.' While AMD’s PEG ratio under 0.5 suggests deep value, this metric is backward-looking and ignores the cyclicality of the PC market and the high R&D costs required to catch up to NVDA’s CUDA software moat. The article glosses over the 'market turmoil' mentioned, failing to address how a high-interest-rate environment impacts the terminal value of these growth-heavy semiconductor stocks.

Devil's Advocate

If the AI infrastructure build-out is truly in its 'early innings' and sovereign AI demand accelerates, NVDA's 75% margins could persist longer than historical hardware cycles suggest, making current valuations look cheap in hindsight.

NVDA, AMD

ChatGPT by OpenAI

▲ Bullish

"Nvidia’s dominant GPU stack and CUDA-driven ecosystem make it the better long-term buy for AI exposure, provided data-center AI spending persists and competitors don’t rapidly close the architectural gap."

Nvidia (NVDA) remains the safer long-term AI play: $68.1B revenue in Q4 FY2026 (+70% YoY) and 75% gross margins reflect both scale and pricing power from GPUs and the CUDA software ecosystem. AMD (AMD) is making tangible CPU share gains (29.2% in Q4 2025) and boasts partnerships with OpenAI/Meta plus an attractive PEG <0.5, but its revenue base ($10.3B Q4 FY2025) and lower margins (54%) mean it needs continued share gains to justify upside. Key missing context: durability of hyperscaler AI spending, unit economics of custom silicon, product cycle timing, and how quickly rivals (Intel, custom ASICs, cloud-designed chips) can erode moats.

Devil's Advocate

Nvidia’s premium multiple is vulnerable if AI capex slows or hyperscalers accelerate bespoke chips; AMD could leapfrog with better performance-per-dollar in data centers, turning its lower valuation into rapid relative outperformance.

NVDA

Grok by xAI

▲ Bullish

"AMD's sub-0.5 PEG and CPU/AI inference exposure make it the superior buy in a semis pullback versus Nvidia's frothier growth narrative."

The article touts Nvidia's explosive 70% YoY revenue growth to $68.1B and 75% gross margins from GPU/CUDA dominance, but AMD's 88% stock return (vs Nvidia's 46%) and CPU market share at 29.2% signal undervaluation—especially with PEG <0.5 vs Nvidia's higher implied multiple. AMD's OpenAI/Meta partnerships and CPU strength position it for AI inference gains, where cost efficiency trumps Nvidia's training moat. In a crash, semis' cyclicality amplifies AMD's re-rating potential if data center ramps continue, though both face softening AI capex risks omitted here.

Devil's Advocate

Nvidia's entrenched CUDA ecosystem and GPU pricing power could sustain its margin lead, leaving AMD as a distant #2 even if CPUs grow.

AMD

The Debate

Claude ▬ Neutral

Responding to Gemini

Disagrees with: Gemini

"ASIC adoption is accelerating but won't crater Nvidia's near-term pricing power; the real battleground is inference margin erosion, which all panelists underweighted."

Gemini flags the ASIC pivot risk correctly, but undersells its timeline. Meta's custom silicon (Trainium, Gaudi) is *already* in production—not theoretical. Yet hyperscalers still buy Nvidia GPUs at scale because software integration and time-to-revenue matter more than per-unit cost. The real question: does AMD's CPU+GPU flexibility let it capture inference workloads faster than Nvidia's installed base locks customers in? Nobody addressed inference margin compression—that's where the 75% gross margin dies first.

Gemini ▬ Neutral

Responding to Claude

Disagrees with: Grok

"AMD's lower gross margins make it more vulnerable than Nvidia to a sudden slowdown in hyperscaler capex."

Claude's focus on inference margin compression is vital, but everyone is ignoring the 'sovereign AI' wildcard. If nations like Saudi Arabia or Japan prioritize energy efficiency over CUDA compatibility, AMD’s EPYC/Instinct integration could win on TCO (Total Cost of Ownership). However, Grok’s focus on AMD's 88% return is a trap; that price action reflects multiple expansion, not earnings reality. If AI capex plateaus, AMD’s lower margins provide a thinner cushion against a valuation collapse than Nvidia’s cash-rich fortress.

ChatGPT ▬ Neutral

Responding to Claude

Disagrees with: Claude

"HBM supply and advanced-node fab allocation are the overlooked bottleneck that could prevent AMD from converting architectural flexibility into meaningful inference revenue gains."

Claude flags inference margin squeeze, but nobody's called out a concrete bottleneck: HBM (high-bandwidth memory) and advanced foundry capacity. If Nvidia retains prioritized access to HBM/leading-node wafers (a plausible, industry-observed dynamic), AMD’s theoretical CPU+GPU advantage can't scale into real-world inference deployments quickly. Add software-porting friction (ROCm vs CUDA) and procurement lead times, and AMD faces a practical ramp constraint the debate underestimates.

Grok ▲ Bullish

Responding to ChatGPT

Disagrees with: ChatGPT

"AMD's HBM/hardware access enables competitive inference scaling, amplifying TCO edges over Nvidia's power-hungry GPUs."

ChatGPT's HBM/foundry bottleneck undersells AMD's MI300X (192GB HBM3E, matching H100) and Samsung partnerships diversifying from TSMC—real constraints are ROCm software lags, not hardware supply. Connects to Claude's inference margins: AMD's CPU+GPU integration slashes power draw 20-30% (per benchmarks), targeting hyperscaler TCO priorities omitted here.