Nvidia and Palantir Are Down 37% and 41% From Their Respective All-Time High -- but It Isn't Time to Buy Just Yet

Though numerous factors played a role in sending the Dow Jones Industrial Average, S&P 500, and Nasdaq Composite to respective closing highs in late 2024 and early 2025, nothing has been more pivotal for the investing community than the evolution of artificial intelligence (AI).

Giving software and systems the capacity to reason and act without the need for human oversight affords AI a seemingly limitless growth runway. It also opens the door for AI-driven software and systems to evolve to learn new tasks without human assistance.

Image source: Getty Images.

In Sizing the Prize, the analysts at PwC estimate AI will grow into a $15.7 trillion addressable global market by 2030.  If this forecast is even remotely in the ballpark, it means countless businesses involved in AI hardware/infrastructure and real-world application are going to benefit.

Arguably no two businesses have been the face of the AI revolution more than graphics processing unit (GPU) giant Nvidia (NVDA -1.64%) and data-mining specialist Palantir Technologies (PLTR -0.62%). At one point, Nvidia tacked on more than $3 trillion in market value and became Wall Street’s most-valuable public company. As for Palantir, its shares peaked in mid-February, which equated to a nearly 2,000% trailing-two-year return.

But how the mighty have fallen.

As of the closing bell on Apr. 4, shares of Nvidia and Palantir had respectively plunged 37% and 41% below their all-time high. While some investors might view this significant downdraft as an opportunity to pounce, three glaring headwinds suggest it’s not time to buy shares of either company just yet.

Next-big-thing bubbles have popped, without fail, for more than three decades

Over the next 10-plus years, there’s a high likelihood AI will transform industries and meaningfully improve the operating efficiency and margins of prominent public companies. But over the short run, history is definitely not in the corner of AI giants like Nvidia and Palantir.

Dating back to the proliferation of the internet in the mid-1990s, every next-big-thing technology has endured an early-stage bubble-bursting event. Without fail, investors consistently overestimate the initial adoption rate and early-stage utility of next-big-thing trends. With expectations overshooting reality, the bubble eventually bursts.

The simple fact that most businesses are nowhere close to optimizing their AI solutions or generating a positive return on their investments in artificial intelligence signals that investors have, once again, overestimated early utility and adoption rates.

If there’s a silver lining here for Nvidia and Palantir, it’s that both benefit from healthy backlogs. Palantir typically signs multiyear contracts with the U.S. government for its Gotham platform, while overwhelming demand for Nvidia’s Hopper and Blackwell GPUs have orders backlogged. Even if the AI bubble bursts, as history suggests it will, sales wouldn’t immediately fall off for either company.

Nevertheless, businesses on the cutting edge of next-big-thing trends are typically hit hardest when bubbles burst. If history were to rhyme once more, shares of Nvidia and Palantir would have considerably further to fall.

Image source: Getty Images.

Tariffs create short-term operating uncertainty

Although most investors are probably tired of hearing about President Donald Trump’s tariff policy, it’s the second big reason of three why investors would be wise to keep their distance from Nvidia and Palantir Technologies.

On April 2nd, a day Trump referred to as “Liberation Day,” he introduced a sweeping 10% global tariff, as well as a flurry of reciprocal tariffs on countries that have historically held adverse trade imbalances with the U.S. While President Trump’s goal is to raise revenue via tariffs, protect American jobs, and return manufacturing back to the U.S., it’s quite possible this approach could have negative consequences.

In December, four New York Federal Reserve economists at Liberty Street Economics published a report that examined the performance of stocks exposed to Trump’s China tariffs in 2018-2019. Unsurprisingly, companies exposed to China tariffs performed worse on days these tariff announcements were made. But what you might not realize is that these underperforming companies also, on average, saw their sales, profits, employment, and labor productivity decline from 2019 to 2021. 

Even though Nvidia doesn’t import from China, it does have contractual agreements in place with chip fabrication giant Taiwan Semiconductor Manufacturing. Countries that are critical to the semiconductor industry were hit with sizable tariffs, which could negatively impact Nvidia’s margins, as well as hurt demand for its products in these overseas markets.

While Palantir doesn’t have to worry about direct tariff implications -- it offers cloud-based, AI- and machine learning-driven platforms -- there is the possibility that worsening trade relations caused by tariffs could reduce demand for the company’s solutions outside the U.S.

Valuation remains a big concern

The final of three reasons that explains why Nvidia and Palantir Technologies aren’t worth buying just yet is their respective valuations.

To state the obvious, value is itself subjective. What one investor finds to be pricey might be viewed as a bargain by another. Nevertheless, history has left little margin for error with either company.

For example, when Palantir reached its all-time closing high of $124.62 per share on Feb. 18, its price-to-sales (P/S) ratio hovered around 100! To put this figure into context, other leading businesses of next-big-thing trends, such as Amazon, Microsoft, and Cisco Systems prior to the dot-com bubble, peaked with P/S ratios ranging from 31 to 43. Even Nvidia’s P/S ratio topped out at 42 last summer. Despite losing 41% of its value, Palantir’s P/S ratio is still hovering around 60, which is a level history conclusively shows isn’t sustainable.

It should be noted that while Nvidia’s P/S ratio has more-than-halved to 18, it’s still considerably pricier than its “Magnificent Seven” peers, relative to sales.

Magnifying this valuation concern is the reality that the stock market is historically pricey. It entered 2025 with the third-highest Shiller price-to-earnings (P/E) Ratio when back-tested to January 1871. Using correlations as a guide, the five previous times the Shiller P/E surpassed 30 during a bull market eventually led to declines of 20% (or greater) in the Dow, S&P 500, and/or Nasdaq Composite.

When the stock market rolls over, companies with premium valuations tend to get hit hard, as you’re probably noticing with Nvidia and Palantir. However, the stock market isn’t close to being inexpensive, despite the recent tumble in equities. With more downside projected by the Shiller P/E Ratio, shares of Nvidia and Palantir are likely to be dragged lower.