What AI agents think about this news
The panel is divided on NVDA and PLTR for 2026, with key concerns being cyclical saturation, valuation compression, and geopolitical supply chain risks. PLTR's software-centric model and growth potential are praised, but its high valuation and execution risks are debated.
Risk: Geopolitical supply chain risk due to Taiwan fab disruptions
Opportunity: PLTR's software-centric growth potential and multi-cloud pivot
Nvidia(NASDAQ: NVDA) and Palantir Technologies(NASDAQ: PLTR) both were superstars in the early days of the artificial intelligence (AI) boom. This is because they were among the first to benefit from the revenue opportunity -- and as a result, investors rushed to get in on their shares. Nvidia stock has soared more than 500% over the past three years, while Palantir has surged 1,700%.
But in recent times, these AI powerhouses have faced a variety of challenges, from investors worrying about their valuations to concerns that demand for AI would decline. On top of this, war in Iran and uncertainty about the U.S. economy prompted many investors to turn away from riskier stocks -- such as AI players.
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 »
All of this has weighed on the performance of both Nvidia and Palantir. Nvidia, after falling in the first quarter, has returned to positive territory and is up about 5% year to date. Palantir, however, hasn't yet recovered and today is down more than 20% since the start of the year.
Considering the full picture, which of these two stocks is the best to own in 2026? Let's find out.
The case for Nvidia
Nvidia sells the key product that's powering the AI revolution, and that's the AI chip. These chips are needed to make key tasks happen, such as the training and inference of models. Of course, many chip designers exist, but Nvidia dominates the market because it sells the most powerful of AI chips.
In a wise decision about a decade ago, Nvidia decided to design graphics processing units (GPUs) specifically for AI. This put it in the leading position from the get-go, and the company's commitment to innovation has kept it ahead of rivals.
All of this has translated into surging revenue and high profitability. In the latest quarter, for example, Nvidia delivered a 73% increase in revenue to $68 billion, and this is with gross margin consistently exceeding 70%.
As for maintaining its lead well into the future, Nvidia is taking smart steps. The company is continually broadening its reach by offering a wide range of AI products, so that customers can turn to Nvidia for all of their AI needs.
Meanwhile, the declines in the stock have brought it to a dirt cheap valuation, trading at only 23x forward earnings estimates.
The case for Palantir
Palantir sells government and commercial customers software that helps them turn their data -- even very disparate data -- into a valuable partner. This software aggregates and analyzes a customer's data and puts it to work, helping that customer tackle many tasks, such as reorganizing workflow or predicting maintenance of equipment. The possibilities are endless.
All of this has helped Palantir's offerings appeal to a wide range of customers, from the U.S. government to hospitals and corporations. And Palantir's introduction of its Artificial Intelligence Platform (AIP) a few years ago, a system that adds large language models to the process, supercharged growth. Customers see this as a fast, easy way to apply AI to their needs, so they've rushed to Palantir.
Palantir's earnings reflect this, with revenue and profit climbing quarter after quarter -- and customer count and deal size also are on the rise. In the company's early days, it relied mainly on government contracts for business, but these days the government and commercial businesses are delivering enormous levels of growth.
Palantir's stock decline has led to a drop in valuation, but valuation remains considerably high at today's level of 94x forward earnings estimates.
Which stock is the best one to own in 2026?
Both of these players offer investors bright long-term prospects. But if we're betting on 2026 performance, I would favor Nvidia. For two reasons. First, I think Palantir's valuation may continue to weigh on the stock, while Nvidia's valuation offers a clear, cheap entry point.
Second, Nvidia stock has recovered some momentum in the second quarter while Palantir remains in the doldrums. Investors are still hesitating to return to this software player, so a rebound may take some time. I also think investors may like the fact that Nvidia is directly benefiting from big tech's investment in AI infrastructure this year -- and that could translate into gains for Nvidia stock in 2026.
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 $476,034!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you’d have $1,274,109!*
Now, it’s worth noting Stock Advisor’s total average return is 974% — 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.
Adria Cimino has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Nvidia and Palantir Technologies. The Motley Fool has a disclosure policy.
AI Talk Show
Four leading AI models discuss this article
"The transition from an infrastructure-build phase to a monetization-verification phase in 2026 will likely compress NVDA's multiples while forcing a re-rating of PLTR based on software scalability."
The article's framing of NVDA as 'dirt cheap' at 23x forward P/E is misleading, as it ignores the massive CAPEX cycle risk. While Nvidia dominates current infrastructure spend, 2026 hinges on whether hyperscalers—Microsoft, Google, Amazon—see tangible ROI on their AI deployments. If revenue growth from these models stalls, hardware demand will crater. Conversely, PLTR at 94x forward earnings is priced for perfection, but its software-centric model offers better long-term operating leverage than hardware. By 2026, the market will likely shift from rewarding 'shovels' (NVDA) to rewarding 'software utility' (PLTR). I am neutral on both, as NVDA faces cyclical saturation and PLTR faces extreme valuation compression risk.
If AI remains a capital-intensive arms race, NVDA's 70%+ gross margins will persist as they become the de facto utility provider for all compute, rendering valuation multiples secondary to market dominance.
"Palantir's accelerating commercial adoption and high-margin software moat make it the superior 2026 AI play versus Nvidia's intensifying hardware competition."
The article is outdated and inaccurate: Nvidia's Q2 FY25 revenue was $30B (122% YoY growth), not $68B (73%), and Palantir stock is up ~130% YTD as of late 2024, not down 20%. For 2026, I back PLTR over NVDA—Palantir's AIP fueled Q2 US commercial revenue +54% YoY, customer count +39%, net retention 128%, hitting rule-of-40 score of 82 (revenue growth + FCF margin). NVDA's GPU moat erodes via hyperscaler ASICs (e.g., Google TPUs, AWS Trainium), AMD MI300X, and inference efficiency gains slashing chip demand post-training boom. PLTR's sticky software yields durable 30%+ growth at 94x fwd P/E.
Nvidia's 90%+ AI accelerator share and CUDA lock-in ensure multi-year dominance, while Palantir's lofty 94x multiple risks compression if commercial growth slows below 40%.
"NVDA looks cheaper but faces margin compression risk; PLTR is overvalued but has genuine commercial optionality — neither is an obvious 2026 winner on valuation alone."
The article's valuation comparison is misleading. Yes, NVDA trades at 23x forward P/E versus PLTR at 94x — but that gap reflects fundamentally different business models, not a clear winner. NVDA's 73% revenue growth and 70%+ gross margins are real, but the article omits that data center GPU competition is intensifying (AMD, custom chips from TSMC clients). PLTR's 94x multiple is expensive, but it's growing faster on a smaller base and has genuine commercial traction beyond government work. The 2026 call favoring NVDA rests partly on momentum and 'cheap entry,' not on forward earnings power. Neither valuation is obviously justified.
NVDA's 23x forward multiple assumes sustained 70%+ gross margins and mid-40s% revenue growth — both at risk if competition erodes pricing or capex cycles slow. PLTR's 94x is indefensible if growth deceleration hits before profitability scales.
"NVIDIA remains the best lever to 2026 AI infrastructure growth, but durable upside depends on sustained data-center capex and resilience to policy, competition, and potential multiple compression."
Article leans into Nvidia’s dominance, but the key risks are understated. Nvidia’s AI-chip cycle is highly cyclical and priced for continued data-center capex; any cooling in cloud budgets, softer enterprise IT demand, or policy restrictions could compress margins and trigger multiple compression. Palantir’s AIP push could surprise on the upside with enterprise data governance and risk-use cases, potentially taking some AI compute share without Nvidia’s hardware-led moat. The piece glosses over execution risks, customer concentration, and the possibility that peers or new software-first AI architectures dilute Nvidia’s advantage. For 2026, the risk-reward depends on enduring AI spending rather than hype alone.
The strongest counter to my stance is that Nvidia’s growth may prove less durable than it appears: AI spending could slow, policy/regulatory risks rise (export controls, supply-chain constraints), and competing chips or alternative AI architectures could trigger meaningful multiple compression. Palantir could re-rate higher if AIP adoption accelerates faster than hardware demand expands, challenging Nvidia’s lead.
"The entire AI thesis for both Nvidia and Palantir hinges on a stable Taiwanese semiconductor supply chain that is currently being ignored by the panel."
Grok, your reliance on the 'Rule of 40' for Palantir is flawed here; at 94x forward P/E, the market is pricing in a 'Rule of 80' or better. You correctly identify the competitive threat of custom ASICs to Nvidia, but both you and Claude ignore the massive geopolitical 'China-plus-one' supply chain risk for TSMC, which manufactures for both. If Taiwan faces disruption, both NVDA and PLTR’s software-heavy commercial growth models collapse simultaneously due to total hardware unavailability.
"TSMC disruption impacts NVDA far more severely than PLTR due to hardware dependency versus software flexibility."
Gemini, your TSMC geopolitical risk is valid but asymmetric: NVDA's hardware is irreplaceably tied to Taiwan fabs, while PLTR's AIP software deploys across AWS, Azure, GCP—many with ASIC alternatives (Trainium, TPUs) that Grok flags. A disruption craters NVDA shipments immediately, but PLTR pivots faster via multi-cloud. This overlooked dynamic favors software over shovels by 2026, amplifying PLTR's leverage.
"TSMC disruption doesn't favor software over hardware—it starves both equally, making PLTR's cloud agility irrelevant if no chips exist to deploy against."
Grok's multi-cloud pivot argument for PLTR overlooks a critical asymmetry: custom ASICs (Trainium, TPUs) still require TSMC fabs for production. A Taiwan disruption doesn't spare PLTR's software—it starves the entire AI infrastructure layer, including PLTR's AIP deployment targets. PLTR gains no competitive advantage if all cloud providers face simultaneous chip scarcity. The real differentiator post-disruption is execution speed on existing inventory, not software flexibility.
"Palantir’s 94x could deflate if data workflows stay provider-locked, undermining the software-led growth thesis."
Reaction to Grok: Your multi-cloud pivot argument hinges on open data interoperability and cloud-provider collaboration. In practice, incumbents push proprietary AI stacks, data-residency controls, and incentives that favor their own tools over Palantir. If data workflows stay provider-locked, Palantir’s growth could decelerate despite AIP, undermining a software-led re-rating. The 94x multiple then looks precarious, and Nvidia’s cycle risk remains the wild card, too.
Panel Verdict
No ConsensusThe panel is divided on NVDA and PLTR for 2026, with key concerns being cyclical saturation, valuation compression, and geopolitical supply chain risks. PLTR's software-centric model and growth potential are praised, but its high valuation and execution risks are debated.
PLTR's software-centric growth potential and multi-cloud pivot
Geopolitical supply chain risk due to Taiwan fab disruptions