By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panelists agree that Nvidia's dominance in AI is vulnerable to hyperscaler custom silicon, which could significantly impact both companies' valuations. While Nvidia's current growth trajectory and CUDA moat are compelling, the risk of commoditization and customer concentration is high. AMD's diversification offers downside protection, but its CPU business may face obsolescence from the same customers it hopes to serve.
Risk: Hyperscaler custom silicon squeezing both Nvidia's and AMD's market share and potentially leading to a significant valuation correction.
Opportunity: AMD's diversification into CPUs and edge markets offering downside protection against Nvidia's potential market share loss.
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 →
Over the last few years, Nvidia (NASDAQ: NVDA) and Advanced Micro Devices (NASDAQ: AMD) have traded higher, primarily due to successes in the AI accelerator market. Over that time, Nvidia has dramatically outperformed AMD as it pioneered this market, while AMD has worked to close that gap.
Nonetheless, that stronger performance goes well beyond the AI accelerator industry. It also relates closely to the surprising differences between the two companies, and that divergence may also explain the differing value propositions between the two semiconductor stocks.
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 »
Admittedly, many of the differences relate to recent changes at Nvidia. As recently as four years ago, when it reported its numbers for the first quarter of fiscal 2023 (ended May 1, 2022), Nvidia was a more diverse company.
At that time, its data center segment, which now designs AI accelerators, had just become the company's largest business segment. The other large segment was gaming, which had traditionally been Nvidia's strength due to its specialty in producing GPUs. Professional visualization and its automotive and robotics segment were also smaller but significant parts of the business.
Fast-forward to the company's first quarter of fiscal 2027 (ended April 26, 2026), and the data center segment now makes up 92% of the company's revenue. Consequently, it no longer breaks out revenue for the other three segments that had existed four years ago, including gaming. Instead, the other 8% of its revenue comes from a segment focused on edge computing, which now includes its gaming products.
Fortunately, with the massive demand for AI chips, this approach has worked out well. Revenue rose by 85% in fiscal Q1 and 65% in fiscal 2026. Considering that the stock trades at just 31 times earnings, it has arguably become too cheap to ignore.
In contrast, instead of focusing heavily on one business, AMD has more explicitly maintained the distinction of its primary business segments. That began when Lisa Su became AMD's CEO in 2014, and she shifted the company's emphasis to CPUs and GPUs.
Hence, when going back four years to the first quarter of 2022, computing and graphics, plus an enterprise, embedded, and semi-custom division, made up nearly all of its revenue.
Those segments changed to their present form one quarter later. However, the data center business, which designs its AI accelerators, is not dominant in the way Nvidia's is. Instead, it made up around 56% of revenue in the first quarter of 2026, with the client and gaming segment claiming 35% of AMD's revenue.
Its embedded segment claims the remaining 9% of revenue. This is not far below where it was when it first purchased Xilinx, indicating it has not abandoned this business.
Nonetheless, the data center business makes one wonder about AMD's direction. Is it going to become primarily focused on AI accelerators like Nvidia, or will it maintain more diversity?
The diversity strategy may be more likely. Since the embedded segment focuses on edge computing and physical AI, the company has good reason to maintain that segment. Additionally, investors should remember that AMD also focuses heavily on CPUs. These have become more critical to data center build-outs and are not a specialty for Nvidia.
That approach paved the way for AMD's 38% annual revenue growth in Q1 and a 34% annual gain in 2025. Indeed, its 161 P/E ratio may deter new investors. Still, even if it does not close the competitive gap with AI accelerators, AMD's diverse approach puts it in a strong position to remain competitive in the AI market.
When comparing Nvidia and AMD, it seems industry dominance, specifically when it comes to AI accelerators, has made Nvidia the standout between the two stocks.
Admittedly, AMD's diverse approach of CPUs, GPUs, and embedded chips that can support AI data centers should serve it well and is arguably the right approach as it seeks to catch up to Nvidia. Also, AMD's revenue growth is such that investors are unlikely to turn against the stock long-term.
However, that does not appear to be enough to overcome Nvidia's first-mover advantage in AI accelerators. Moreover, Nvidia offers faster growth and a much lower valuation, which appears to make the stock a better buy at this time.
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 $371,842! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,244,783!
Now, it’s worth noting Stock Advisor’s total average return is 900% — a market-crushing outperformance compared to 207% 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 July 19, 2026. *
Will Healy has positions in Advanced Micro Devices. The Motley Fool has positions in and recommends Advanced Micro Devices 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.
Four leading AI models discuss this article
"AMD's diversified CPU/GPU/embedded strategy offers a hedge against Nvidia's extreme concentration risk that the article underplays, making a binary 'NVDA is clearly better' conclusion premature at current valuations."
The article correctly notes Nvidia's extreme concentration (92% data center revenue) versus AMD's more balanced mix (56% data center, 35% client/gaming, 9% embedded). NVDA's 85% Q1 growth and 31x P/E look compelling against AMD's 38% growth at 161x P/E. However, the piece glosses over Nvidia's massive customer concentration risk (primarily hyperscalers racing to build AI), potential margin compression as CUDA moat erodes, and AMD's CPU share gains in data centers plus its Xilinx-derived edge/embedded portfolio that could capture physical AI and inference markets Nvidia is de-emphasizing. Missing context: AMD's EPYC ramp and MI300X traction, while slower, are real and diversify beyond pure accelerator bets.
If Nvidia successfully productizes its full-stack software+hardware advantage into sovereign AI and enterprise deployments while AMD's catch-up in accelerators stalls, the valuation gap will widen further and AMD's diversification could simply mean slower growth and lower returns on capital.
"AMD’s high P/E is a temporary accounting artifact of recent acquisitions and R&D, masking a superior long-term architectural flexibility for the inference-heavy phase of AI deployment."
The article's valuation comparison is misleading. Comparing Nvidia’s 31x forward P/E to AMD’s 161x is a classic apples-to-oranges trap; AMD’s current GAAP earnings are heavily suppressed by amortization from the Xilinx acquisition and aggressive R&D spend. Nvidia is currently a pure-play AI utility, while AMD is a diversified compute platform. The real risk isn't just 'market share'—it's software ecosystem lock-in. Nvidia’s CUDA moat is the true barrier, not just hardware throughput. If the AI market shifts from training (Nvidia's stronghold) to inference (where CPU/GPU balance matters more), AMD’s 'diverse' architecture could actually become a massive structural advantage, potentially leading to a significant valuation re-rating as margins normalize.
If the AI arms race hits a capital expenditure wall, Nvidia's 92% data center reliance becomes a liability, whereas AMD’s legacy CPU business provides a necessary floor for cash flow.
"The article mistakes valuation compression for valuation opportunity; Nvidia's lower multiple reflects priced-in competition risk, not a bargain, while AMD's higher multiple compensates for execution risk in a narrower AI accelerator market share."
The article's core claim—that Nvidia's 31x P/E is 'too cheap to ignore' while AMD's 161x P/E deters investors—inverts reality. Nvidia trades at a 5x multiple discount despite 85% YoY revenue growth and 92% data center concentration because the market prices in: (1) AI accelerator commoditization risk as competition intensifies, (2) customer concentration (likely hyperscalers), and (3) the reality that 31x on $60B+ revenue is not cheap by historical standards. AMD's higher multiple reflects optionality: CPU leverage in AI infrastructure, embedded/edge diversification, and lower single-point-of-failure risk. The article conflates 'first-mover advantage' with 'durable moat' without examining whether Nvidia's dominance survives a 2-3 year competitive cycle.
If Nvidia's data center TAM expands faster than AMD can scale, and if hyperscalers lock in CUDA ecosystem switching costs, then Nvidia's concentration becomes a feature, not a bug—and its 31x multiple could re-rate upward while AMD's diversity dilutes returns.
"NVIDIA’s AI data-center cycle, supported by a software moat and hyperscale demand, should drive multi-year earnings resilience even if short-term multiples compress."
The article frames Nvidia as unstoppable in AI and AMD as the diversified challenger, but it understates Nvidia’s near-term vulnerability to AI capex cycles: a meaningful slowdown or a shift to alternative accelerators or open software could compress margins and the stock multiple given data-center revenue concentration. It also glosses over the CUDA software moat’s fragility if ecosystems pivot to open stacks or regulatory/geopolitical frictions hit hyperscale spend. AMD’s breadth helps, but it doesn’t guarantee parity with Nvidia’s AI-accelerator edge. In the near term, Nvidia looks compelling, but valuation and cycle risk keep upside and downside in play.
Strongest counter: Nvidia’s CUDA/software moat and entrenched hyperscale demand could sustain growth longer than the article assumes, making the AI-cycle risk a bigger threat for AMD than for Nvidia.
"Nvidia's 31x forward P/E remains attractive relative to its projected growth and margin durability versus AMD's optically cheaper diversification."
Claude's claim that Nvidia's 31x P/E already prices in commoditization and customer concentration ignores the forward growth trajectory. With consensus expecting ~50% revenue CAGR through FY27 and gross margins holding above 75%, the multiple implies only modest re-rating even if share erodes 5-10pts. AMD's 'optionality' in CPUs and edge still lacks comparable AI software traction, keeping its 161x multiple extended.
"Hyperscaler-developed custom silicon poses a greater threat to AMD's CPU-led diversification than the market currently prices into either stock."
Claude, you’re mispricing the 'optionality' of AMD’s CPU business. That isn't a growth engine; it's a defensive anchor in a server market currently cannibalized by AI spend. While Gemini highlights inference as a potential AMD stronghold, the industry is moving toward custom silicon (ASICs) from hyperscalers like Google and Amazon. This creates a squeeze: Nvidia faces commoditization, but AMD faces obsolescence from the very customers it hopes to serve. Both valuations are ignoring the looming capex efficiency correction.
"Custom silicon adoption by hyperscalers poses an existential threat to both Nvidia and AMD's growth assumptions that current multiples don't price in."
Gemini's custom silicon point cuts deeper than anyone acknowledged. Google TPUs, Amazon Trainium, and Meta's custom accelerators aren't theoretical—they're already shipping. This doesn't just squeeze AMD; it fundamentally reshapes Nvidia's TAM. If hyperscalers shift 20-30% of workloads to proprietary ASICs over 3 years, Nvidia's 50% CAGR assumption collapses regardless of CUDA moat. Neither valuation accounts for this structural shift.
"Hyperscaler custom ASICs could shrink Nvidia’s TAM and pricing power faster than current expectations."
Claude’s focus on the 31x P/E framing misses a structural risk: hyperscaler custom silicon. If 20–30% of workloads migrate to private ASICs over 3 years, Nvidia’s TAM and pricing power could compress faster than consensus expects, even with CUDA moat. That implies multiple re-rating risk beyond the current cycle, and AMD’s diversification may offer downside protection to Nvidia’s growth story. We should price in that scenario.
The panelists agree that Nvidia's dominance in AI is vulnerable to hyperscaler custom silicon, which could significantly impact both companies' valuations. While Nvidia's current growth trajectory and CUDA moat are compelling, the risk of commoditization and customer concentration is high. AMD's diversification offers downside protection, but its CPU business may face obsolescence from the same customers it hopes to serve.
AMD's diversification into CPUs and edge markets offering downside protection against Nvidia's potential market share loss.
Hyperscaler custom silicon squeezing both Nvidia's and AMD's market share and potentially leading to a significant valuation correction.