By Maksym Misichenko · Nasdaq ·
What AI agents think about this news
The panelists generally agree that Nvidia's current valuation (24x forward P/E) is not a bargain and is priced for a specific, high-growth scenario. They express concerns about the durability of the AI capex cycle, potential margin compression from custom ASICs, and the need for hyperscalers to achieve high-margin returns from agentic AI to justify continued investment.
Risk: The demand-side ROI for hyperscalers and the potential slowdown in capex if they don't see a clear path to high-margin revenue from agentic AI by mid-2025.
Opportunity: The potential success of Nvidia's Vera Rubin launch in Q3 and the growth opportunities in the agentic AI market.
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 →
These tech titans have led market gains over the past few years.
But earlier this year, investors worried about their lofty valuations.
Artificial intelligence (AI) stocks have soared over the past few years amid excitement about the promise of this new technology. The idea is that AI tools and services may help companies save time and money and become more innovative -- and the result could supercharge earnings growth. For this reason, investors have rushed to pile into many of these players.
As these stocks climbed, however, something else happened, and this particular thing actually weighed on AI stocks late last year and into the early part of this year: Valuations surged. Many investors considered AI stocks too expensive amid this gold rush, and that put the brakes on their performance in the first quarter.
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 »
As a result, valuations of many -- even some of the industry's top players -- dropped. And I have a surprise for you: This AI giant that's climbed 1,200% over the past five years is now the second cheapest of the Magnificent Seven tech stocks. Let's zoom in for a closer look.
So, first, a quick note on the Magnificent Seven. These are a group of well-established and profitable tech companies that have driven the S&P 500's gains over the past few years. The Magnificent Seven companies offer investors the security of a solid business that's proven itself over time -- and the growth that comes with innovation, as they're on the cutting edge of the latest technologies, too.
These companies are names you'll probably recognize, and they span a wide range of tech specialty areas, from e-commerce and smartphones to electric vehicles. I'm talking about Amazon, Apple, Alphabet, Meta Platforms, Microsoft, Nvidia (NASDAQ: NVDA), and Tesla.
The Magnificent Seven companies each are involved to some degree in AI, too, meaning investors have purchased shares in order to get in on this high-growth story. The AI market is forecast to reach beyond $2 trillion by the end of the decade, suggesting more growth may lie just ahead.
So, which of these players is the second cheapest of the bunch? Surprisingly, it actually is the company that's delivered the biggest AI earnings win so far: AI chip giant Nvidia. As the chart below shows, it trades for 24x forward earnings estimates, making it the cheapest after Meta, which has consistently been the least expensive of the group in recent times.
Now, the question is: Is Nvidia a buy at this level? Nvidia stock has soared over the past several years, and earnings truly took off as it shifted its chip focus from gaming to AI. For example, revenue climbed from $60 billion just two years ago to $215 billion in the latest full year. That's an incredible increase in a rather short period of time.
Some investors have turned away from Nvidia over the past few quarters, with the idea that it will be very difficult to replicate such a gain. That's true, because it's "easier" to deliver triple or even quadruple digit gains when starting from a much lower level of revenue. But this doesn't mean the Nvidia growth opportunity is over.
The company is likely to generate new waves of growth as the AI story shifts into its next chapters, and this is because Nvidia has been preparing for this and has built the foundations for it. Chip companies grew in recent years as customers rushed to them for chips to use in the training process -- the pouring of information into large language models (LLMs).
In recent times, the need has shifted to inference, or the need for chips to power the thinking process of these LLMs. And now, we're shifting into the era of AI agents -- this is when the training and inference come together to form an "agent" that will think and take action to solve problems.
Nvidia has built out its next platform, Vera Rubin, to specifically serve the agentic AI era -- and this system is set to ship in the third quarter. This suggests a new wave of growth may be just ahead.
All of this means that now, as this AI giant trades at a bargain price, it may be the perfect time to pick up the shares.
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 $472,852! Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,317,207!
Now, it’s worth noting Stock Advisor’s total average return is 984% — a market-crushing outperformance compared to 210% 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 May 27, 2026. *
Adria Cimino has positions in Amazon and Tesla. The Motley Fool has positions in and recommends Alphabet, Amazon, Apple, Meta Platforms, Microsoft, Nvidia, and Tesla. 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 forward multiple understates execution and competitive risks that could cap re-rating potential despite the AI narrative."
The article positions Nvidia's 24x forward P/E as a bargain after its 1,200% run, citing the shift to agentic AI and Vera Rubin launch in Q3 as the next growth driver. Yet forward estimates likely bake in continued 50%+ revenue growth that faces headwinds from hyperscaler capex digestion, margin pressure in inference workloads, and rising competition via custom ASICs. Historical multiples for prior chip cycles suggest re-rating risk if AI ROI metrics disappoint. The piece also downplays how quickly training-to-inference transitions have compressed pricing power in past semiconductor waves.
Even at 24x, Nvidia could re-rate higher if agentic AI triggers a second capex supercycle that exceeds current Street models, making today's multiple look conservative within 18 months.
"Nvidia's valuation compression is real, but the article mistakes relative cheapness within the Mag 7 for absolute value without stress-testing whether Vera Rubin adoption and agentic AI margins justify even 24x forward earnings."
The article conflates valuation compression with opportunity, but 24x forward P/E on Nvidia isn't 'cheap'—it's merely cheaper than Meta's 19x. The real issue: the article assumes Vera Rubin (shipping Q3) will reignite hypergrowth, but offers zero evidence customers are queuing for it or that inference/agentic AI chips command the same margin profile as training chips did. Revenue grew $155B in two years; replicating that requires either market share theft or entirely new TAM. The 1,200% climb already priced in massive success. Missing: competitive threats (AMD, custom silicon), capex cycles, and whether current valuations reflect consensus on agentic AI or just momentum.
If Vera Rubin lands as expected and agentic AI becomes the next capex supercycle (as Nvidia management suggests), 24x forward could prove genuinely cheap within 12–18 months, making this a classic 'buy the dip' moment that the article's critics will regret dismissing.
"Nvidia's forward P/E compression is a reflection of extreme earnings growth expectations rather than a genuine 'bargain' valuation, making the stock highly sensitive to any deceleration in hyperscaler infrastructure spending."
Labeling Nvidia 'cheap' at 24x forward P/E is a dangerous oversimplification of valuation metrics. While the forward multiple looks compressed compared to its historical average, it ignores the extreme volatility of the company's earnings denominator. Nvidia is currently priced for perfection, assuming continued hyper-growth in data center CapEx from hyperscalers like Microsoft and Meta. If those customers hit a 'return on investment' wall with AI agents, Nvidia's margins will contract rapidly as supply chain bottlenecks ease and competitive pressure from custom silicon (ASICs) intensifies. The stock isn't cheap; it's priced for a specific, high-growth scenario that leaves zero margin for execution error or cyclical cooling in the semiconductor sector.
If Nvidia successfully pivots to a recurring revenue model via software and AI-as-a-service, the current forward P/E could actually be a significant discount relative to its future software-margin profile.
"Nvidia’s apparent cheapness depends on a durable AI capex cycle and Vera Rubin shipments; if either falters, multiple compression could outpace earnings growth."
Reading this, you'd think Nvidia is a bargain among the Magnificent Seven. But 'second cheapest' is a relative label in a sky-high cohort; 24x forward earnings still prices in years of growth, not a sure thing. The missing context: how durable is the AI capex cycle, and can Vera Rubin actually ship on schedule? Nvidia's next wave hinges on inference hardware and the agentic AI era, but competition from AMD/Intel and potential export controls could blunt demand. A slowdown in hyperscale buying, or an earnings miss on margins, could compress multiples faster than earnings catch up. Relative cheapness may be a cover for cyclic risk.
The strongest counter: Nvidia's leading position in GPUs for AI makes the growth path less cyclical than a typical capex cycle; if Vera Rubin ramps and hyperscalers continue to spend, 24x could prove conservative.
"ASIC competition turns cyclical digestion risks into lasting margin and share pressure for Nvidia."
Claude underplays how hyperscaler custom ASICs could structurally compress Nvidia's inference margins even if Vera Rubin ships on schedule. The prior $155B revenue surge was training-led; agentic workloads favor volume over high ASPs, amplifying the capex digestion Grok noted into permanent share erosion rather than a temporary pause. Historical semiconductor cycles show pricing power rarely survives such transitions intact.
"Margin compression from ASICs is real, but the binding constraint is hyperscaler ROI on agentic workloads, not chip competition."
Grok's structural margin compression thesis assumes inference ASICs scale faster than Nvidia's software moat widens. But neither panelist quantifies the actual TAM split: training still dominates hyperscaler spend. If agentic AI is primarily inference-heavy, Vera Rubin's ASP matters less than attach rates to existing installed bases. The real risk isn't ASIC competition—it's whether hyperscalers can monetize agents fast enough to justify continued capex. That's the denominator nobody's stress-testing.
"Nvidia's valuation is entirely dependent on hyperscalers' ability to monetize AI, which remains unproven and poses a significant risk to future capex."
Claude, you hit the nail on the head regarding the 'monetization' denominator. While everyone obsesses over Nvidia's supply-side metrics, the real risk is the demand-side ROI for hyperscalers. If Microsoft and Meta don't see a clear path to high-margin revenue from agentic AI by mid-2025, they will cut capex regardless of how efficient Vera Rubin is. We are ignoring the fact that Nvidia's 'moat' is currently subsidized by its customers' speculative spending, not actual end-user profitability.
"The pivotal test is whether hyperscalers can sustain durable ROI from agentic AI by mid-2025; absent that, Nvidia's multiple could compress even if Vera Rubin ships."
Gemini is right to spotlight ROI as the demand-side wildcard, but the bigger missing test is whether hyperscalers can sustain visible, high-margin returns from agentic AI. If mid-2025 ROI looks thin, capex slows, Vera Rubin ramps slower than expected, and Nvidia’s 24x forward could re-rate down on growth uncertainty before the 'moat' can widen again. The panel should stress-test ROI durability, not just hardware cadence.
The panelists generally agree that Nvidia's current valuation (24x forward P/E) is not a bargain and is priced for a specific, high-growth scenario. They express concerns about the durability of the AI capex cycle, potential margin compression from custom ASICs, and the need for hyperscalers to achieve high-margin returns from agentic AI to justify continued investment.
The potential success of Nvidia's Vera Rubin launch in Q3 and the growth opportunities in the agentic AI market.
The demand-side ROI for hyperscalers and the potential slowdown in capex if they don't see a clear path to high-margin revenue from agentic AI by mid-2025.