By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panel is largely bearish on Nvidia's expansion into the CPU market, citing entrenched x86 ecosystems, software compatibility issues, and potential margin compression. They also raise concerns about execution risks, supply chain constraints, and the risk of custom silicon acceleration by hyperscalers.
Risk: Potential margin compression and custom silicon acceleration by hyperscalers
Opportunity: Potential for Nvidia to force a proprietary 'Nvidia-only' stack, making x86 compatibility irrelevant for AI-native data centers
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 →
Nvidia (NASDAQ: NVDA) has been one of the biggest artificial intelligence (AI) success stories so far. The company provides a crucial tool -- and one of the highest quality -- used in the development of this technology. This is the graphics processing unit (GPU), a chip that powers important tasks such as the training of models.
The company's GPU strengths and its portfolio of related products and services have helped it to report record levels of earnings quarter after quarter. And this has lifted the stock too, with gains of more than 400% over three years.
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 »
Some investors have worried that, after such a performance, Nvidia may lose momentum. It's true that there are plenty of rivals in the AI chip space, from chip designers like Advanced Micro Devices to some of Nvidia's customers, like Amazon, that have created their own chips.
But my prediction is Nvidia will stay ahead of the crowd -- and the second half of this year actually will represent a game-changing moment for the AI giant. Let's take a closer look.
So, first, a bit of background on this market leader and where it stands in today's AI environment. Nvidia's GPUs have been around for decades, and in their early days, they mainly served the gaming market. The company has since expanded their use, and this was made possible by Nvidia's creation of CUDA, a parallel computing platform.
And about a decade ago, recognizing the AI opportunity, Nvidia tailored its GPUs for this industry. This, along with the creation of other products to support the GPU in its AI tasks, helped Nvidia build an AI empire. In the latest quarter, the company reported an 85% increase in revenue to more than $81 billion. And gross margin has remained pretty consistently above 70%, showing high profitability on sales.
As mentioned, Nvidia isn't alone in the space. Rivals sell GPUs or other similar AI chips, and they, too, have delivered significant growth. Yet Nvidia has maintained its lead, due to its brand strength and the quality of its products, as well as its focus on innovation.
But some investors have wondered how long this will last, particularly as rivals too have been supercharging their innovation engines -- and Nvidia's GPUs carry the highest price tag. Meanwhile, the needs of AI are changing. For example, the early stage of the AI story was all about training models, and for this, the GPU was critical.
Today, we're moving into the era of AI agents, involving the actual application of AI to problems. In agentic AI, the AI agent acts as a human would -- considering a problem and taking steps, in many cases multiple steps, to solve it. And to power this process, another type of chip is most needed: the central processing unit (CPU). These are the general chips found in all computers.
Nvidia hasn't been a big player in the CPU market. Intel and AMD have been longtime leaders in this market, but if Nvidia meets its goals, this might change.
And this leads me to my prediction. The second half of the year could be a key moment for Nvidia because it plans to take two game-changing steps: It aims to release its Vera Rubin platform for data centers, and this includes the company's first-ever stand-alone CPU. And, for the PC market, it aims to release a new superchip, the Nvidia RTX Spark. This chip, including an Nvidia GPU and an Nvidia CPU, will launch in Windows laptops this fall from Microsoft, Dell, and others.
So, as of the second half, Nvidia will advance in the CPU market in a big way -- aiming for share in data center CPUs and in the PC market. Nvidia says the stand-alone CPU market is worth about $200 billion, and the company says it's on track for leadership.
The big news here is that Nvidia is maintaining its GPU dominance and eventually may hold a similar position in the broader CPU market. This could greatly increase the company's revenue growth potential over time -- and that's why I predict that the launches of these two CPU products will represent a game-changing moment in the Nvidia story.
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 $438,283! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,257,427!
Now, it’s worth noting Stock Advisor’s total average return is 938% — a market-crushing outperformance compared to 206% 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 June 13, 2026. *
Adria Cimino has positions in Amazon. The Motley Fool has positions in and recommends Advanced Micro Devices, Amazon, Intel, Microsoft, 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
"Nvidia's CPU push faces deeper ecosystem and compatibility hurdles than the article acknowledges, limiting near-term revenue impact."
The article positions Nvidia's planned H2 2026 Vera Rubin data-center CPU and RTX Spark PC superchip as a major expansion into the $200B CPU market. Yet Nvidia's existing Arm-based Grace CPU has achieved only niche traction against entrenched x86 ecosystems from Intel and AMD. Hyperscalers continue designing custom silicon, and software compatibility plus ecosystem lock-in remain formidable barriers. Gross margins above 70% are impressive, but entering lower-margin CPU segments could dilute returns if volume gains prove slower than projected.
Nvidia could replicate its CUDA moat in CPUs via superior integration with its GPUs, rapidly capturing share in AI inference workloads where general-purpose chips underperform.
"The article oversells a speculative CPU/PC strategy as a near-term, margin- and revenue-accelerating catalyst; execution risk and timing could prevent a material upside in H2 2026."
Bold claims in the piece hinge on a $200B TAM for data-center CPUs and a Windows-laptop 'RTX Spark' launch, but the numbers in the write-up look dubious. The reference to an 85% revenue jump to more than $81B in a single quarter contradicts Nvidia's known results and likely reflects a misprint or annualized framing. Even if Vera Rubin and the RTX Spark initiative materialize, Nvidia faces execution risk (new CPUs, software, ecosystem), potential margin compression versus GPUs, and competitive pushback from AMD/Intel. The article glosses timing, supply chains, and OEM acceptance—crucial factors that could delay or dilute any so-called game-changing upside.
Counterpoint: if Vera Rubin and RTX Spark achieve rapid OEM adoption and deliver clear data-center and notebook economics advantages, Nvidia could extend its growth and margin upcycle. That would make the cited 'game-changing' moment credible rather than just hype.
"Nvidia's expansion into the CPU market risks compressing their industry-leading margins by forcing them to compete in a commoditized, legacy-entrenched space where their GPU-driven software moat is significantly weaker."
The article’s pivot to the 'CPU era' for Nvidia is a classic example of hardware-vendor hubris. While the Rubin platform and ARM-based CPU entry are technically impressive, the author ignores the massive moat Intel and AMD have built through x86 ecosystem compatibility and legacy enterprise software stacks. Nvidia’s 70%+ gross margins are currently fueled by a supply-constrained GPU monopoly that is already seeing competitive pressure from custom silicon at hyperscalers like Amazon and Google. Pivoting into the commoditized CPU market risks diluting their premium brand and inviting a margin-crushing price war that Nvidia is currently unequipped to win against entrenched incumbents.
If Nvidia successfully leverages its CUDA software ecosystem to make its CPUs the default choice for AI-integrated data centers, they could render the x86 legacy advantage obsolete, effectively forcing a total industry migration.
"Nvidia's H2 2026 CPU launches are real but represent an unproven entry into a mature, competitive market where Nvidia's GPU moat provides minimal advantage."
The article conflates two distinct narratives without evidence connecting them. Yes, Nvidia's H2 2026 CPU launches (Vera Rubin, RTX Spark) are real product events. But the leap from 'Nvidia enters CPUs' to 'game-changing moment' ignores brutal realities: Intel and AMD own $200B in installed CPU base with decades of software optimization, supply chain, and OEM relationships. Nvidia's GPU dominance doesn't transfer to CPUs—different architectures, different moats. The article also cherry-picks 85% YoY revenue growth without noting the law of large numbers: $81B quarterly revenue means incremental growth rates will decelerate regardless of product launches. Finally, 'agentic AI needs CPUs' is speculative; inference workloads still favor GPUs for LLM tasks.
If Nvidia's software stack (CUDA ecosystem) successfully abstracts away CPU architecture differences, and enterprise customers value unified GPU+CPU purchasing from one vendor, Nvidia could capture meaningful share faster than historical CPU transitions suggest—especially in AI-optimized data center workloads where legacy x86 optimization matters less.
"Foundry capacity allocation creates an unmentioned volume and margin trade-off that could delay or shrink Nvidia's CPU opportunity."
ChatGPT flags execution and supply-chain gaps but misses how TSMC's CoWoS and advanced-node capacity, already rationed for Nvidia GPUs through 2025, will force direct trade-offs with Vera Rubin CPU volumes. Any CPU ramp therefore risks cannibalizing higher-margin GPU output or inviting further custom-silicon acceleration at Microsoft and Google before ecosystem shifts can occur.
"Capacity constraints (CoWoS/advanced-node packaging and limited foundry allocations) will throttle Vera Rubin's ramp, preventing Nvidia from translating GPU margins into a meaningful CPU up-cycle on the timelines the article suggests."
Grok overemphasizes capacity harmony between Vera Rubin and GPU sales; in reality, Nvidia's CoWoS/advanced-node packaging and limited foundry capacity will throttle CPU ramp, forcing trade-offs with GPU volumes and delaying margin uplift. Even if Vera Rubin scales, hyperscalers may still prefer in-house CPUs or ARM alternatives, and GPU margins won't automatically transfer to CPUs. This capacity constraint risks a slower, less transformative data-center push than the article implies.
"Nvidia may bypass traditional CPU competition by reframing the CPU as a subordinate control plane for AI-native GPU clusters."
Grok and ChatGPT are fixated on foundry capacity constraints, but they ignore the software-defined silicon reality. If Nvidia’s Vera Rubin architecture effectively treats the CPU as a mere control plane for the GPU, the 'CPU market' definition changes. The risk isn't just supply-side cannibalization; it is the potential for Nvidia to force a proprietary 'Nvidia-only' stack that makes x86 compatibility irrelevant for AI-native data centers, rendering the current Intel/AMD moat a legacy relic rather than a competitive barrier.
"Vera Rubin as GPU-dependent control plane avoids margin compression but forfeits the CPU market claim—and accelerates hyperscaler custom-silicon adoption as defensive hedging."
Gemini's 'control plane' framing is seductive but inverts the real risk. If Vera Rubin becomes GPU-dependent infrastructure rather than a standalone CPU, Nvidia hasn't entered the CPU market—it's just repackaged its GPU moat. That's not margin-dilutive, but it also doesn't justify the '$200B TAM' narrative the article leans on. The actual threat: hyperscalers see this coming and accelerate custom silicon *before* lock-in occurs. Nobody's flagged the timing race between Nvidia's ecosystem lock and customer defection.
The panel is largely bearish on Nvidia's expansion into the CPU market, citing entrenched x86 ecosystems, software compatibility issues, and potential margin compression. They also raise concerns about execution risks, supply chain constraints, and the risk of custom silicon acceleration by hyperscalers.
Potential for Nvidia to force a proprietary 'Nvidia-only' stack, making x86 compatibility irrelevant for AI-native data centers
Potential margin compression and custom silicon acceleration by hyperscalers