No next-big-thing investment trend has arguably been hotter since the advent of the internet than artificial intelligence (AI) is right now.
When discussing "AI," I'm referring to software and systems handling tasks that would normally be overseen by humans. Machine learning gives software and systems the ability to learn and evolve without human intervention, which can make these systems more proficient at their tasks over time.
The utility of AI can't be overstated enough. Almost every sector and industry can benefit from its use, which is probably why the analysts at PwC believe artificial intelligence will add $15.7 trillion to the global economy come 2030.
Image source: Getty Images.
While most institutional money managers and Wall Street analysts view AI and the companies deploying AI infrastructure and software favorably, this bullishness isn't universal. Based on the low-water price targets issued by a select group of Wall Street analysts, the following five AI stocks could plunge by as much as 86%!
Nvidia: Implied downside of 30%
The first AI stock that could tumble after a monstrous run higher is none other than the infrastructure backbone of the AI revolution, Nvidia (NVDA 2.94%). In February, D.A. Davidson analyst Gil Luria raised his and his firms' price target on Nvidia to $620 from $410. Despite this epic increase, it implies that the third-largest public company by market cap in the U.S. will shed 30% of its value, based on where it closed on April 5.
Although Nvidia's A100 and H100 graphics processing units (GPUs) have been the overwhelming top choice of businesses for AI-accelerated data centers, there are plenty of reasons to be believe Nvidia stock could be in a bubble.
To start with, Nvidia's top four customers, which comprise about 40% of its sales, are all developing AI GPUs of their own. Even if these companies continue to rely on Nvidia's GPUs as complements to their in-house GPUs, future orders from these industry leaders would be expected to decline.
To add to the above, Nvidia's pricing power on its A100 and H100 GPUs is expected to take a hit in the latter-half of 2024. The lion's share of its 217% increase in data center sales in fiscal 2024 (ended Jan. 28, 2024) was driven by GPU scarcity. As new competitors enter the arena and Nvidia ramps up its own production, GPU scarcity will diminish -- as will the company's pricing power.
Further, every next-big-thing trend over the last 30 years has navigated its way through an early stage bubble. Suffice it to say, I fully expect Luria's price target to be hit.
NASDAQ: SMCI
Key Data Points
Super Micro Computer: Implied downside of 74%
A second artificial intelligence stock that could plummet in the not-too-distant future if one Wall Street analyst is correct is server and storage solutions company Super Micro Computer (SMCI 2.68%). Senior equity research analyst Mehdi Hosseini of Susquehanna stood firm on his price target of $250 for Super Micro just two months earlier in an interview on CNBC. This would imply Super Micro Computer losing almost three-quarters of its current value.
It's understandable why investors have gone head over heels for Super Micro. The company incorporates Nvidia's top-tier GPUs in its energy-efficient and highly customizable rack servers that are utilized in AI-accelerated data centers. Sales for the company are expected to double in 2024 to north of $14 billion.
However, we've been here before with Super Micro. It was expected to be a prime beneficiary of the cloud-computing revolution in the late 2010s, but failed to maintain its sales momentum. This speaks to the fact that investors have a terrible habit of overestimating the adoption of new trends and innovations.
Super Micro Computer also finds itself at the mercy of Nvidia. If the latter continues to deal with supply issues/scarcity for its GPUs, Super Micro will fail to realize its full potential.
While a retracement to $250 may not be in the cards anytime soon, Super Micro Computer definitely has a lot to prove after its stock appreciated by 767% over the trailing year.
A Tesla Model S charging. Image source: Tesla.
Tesla: Implied downside of 86%
The potential disaster du jour among AI stocks is North America's leading electric-vehicle (EV) manufacturer Tesla (TSLA 2.07%). CEO and founder Gordon Johnson of GLJ Research believes Tesla will retrace to $23.53 per share, which would represent 86% downside from where it closed on April 5.
On one hand, Tesla has made history by becoming the first automotive company to build itself from the ground up to mass production in well over a half-century. It's also the only EV pure-play that's generating a recurring profit (four consecutive years, as of the end of 2023).
At the same time, Tesla is encountering game-changing challenges as EV demand tapers off. The company slashed the sales price of its four production models (3, S, X, and Y) on more than a half-dozen occasions in 2023. The end result has been a more-than-halving in Tesla's operating margin since the end of September 2022 (17.2% to 8.2%, as of Dec. 31, 2023).
Perhaps an even bigger issue is that Tesla has failed to become more than just a car company. Its Energy Generation and Storage segment sales have stagnated in recent quarters, while Services is generating a low-single-digit gross margin. I'll add that solar has been a money-loser since SolarCity was acquired in 2016.
Despite being nothing more than a car company, Tesla is valued at nearly 60 times forecast earnings in 2024. With most auto stocks trading in a range of 6 to 8 times expected earnings per share (EPS), there's definite reason to believe Tesla stock will head lower.
NASDAQ: MSTR
Key Data Points
MicroStrategy: Implied downside of 85%
A fourth artificial intelligence stock that could get absolutely crushed if the most-bearish Wall Street analyst proves accurate is enterprise analytics software company MicroStrategy (MSTR 4.46%). Last August, Jefferies analyst Brent Thill defended his firms' lowball price target of $210 for MicroStrategy, which would imply a decline of 85% from where shares closed this past week.
One issue for MicroStrategy is that its AI-driven enterprise analytics software division stopped growing 10 years ago. Sales have fallen by an aggregate of 14% since peaking.
But let's face the facts: MicroStrategy is best-known for CEO Michael Saylor's strategy of issuing convertible debt and using the proceeds to buy Bitcoin (BTC +0.27%), the largest cryptocurrency by market cap. On March 19, Saylor made a post on X (the platform formerly known as Twitter) that noted his company held 214,246 Bitcoin -- more than 1% of the eventual supply -- at an average price of roughly $35,160 per token.
With each Bitcoin valued at $67,600, as of the time of this writing, MicroStrategy's crypto stake is worth $14.5 billion. However, the company's market cap currently stands at $24.4 billion. Accounting for the company's modestly profitable (but non-growing) software business, as well as its growing debt pile, MicroStrategy is trading at around a 70% premium to the Bitcoin it holds.
If investors want to buy Bitcoin, a spot Bitcoin exchange-traded fund (ETF) or a direct purchase of the cryptocurrency on an exchange would be a considerably smarter move than buying into MicroStrategy, whose premium valuation to the current price of Bitcoin makes no sense.
NASDAQ: PLTR
Key Data Points
Palantir Technologies: Implied downside of 78%
The fifth AI stock that could plunge, according to the prognostication of one Wall Street analyst, is data-mining company Palantir Technologies (PLTR +3.03%). Analyst Rishi Jaluria of RBC Capital doesn't believe Palantir's Commercial segment margins can continue expanding at their current pace. This compelled Jaluria to recently reiterate his and his firms' $5 price target on Palantir, which implies 78% downside.
There's no question Palantir's stock is pricey. The company's nearly $51 billion market cap is closing in on a multiple of 19 times forecast sales this year. It's also valued at 70 times consensus EPS in 2024 despite slowing sales growth.
But one thing Palantir has working for it that merits a hefty premium is its irreplaceability. No company at scale comes anywhere close to the solutions it can provide. Palantir's AI-driven Gotham platform helps governments cull data and plan military operations. The contracts Palantir wins from the U.S. government tend to span multiple years and generate highly predictable cash flow.
Meanwhile, its Foundry platform is just getting off the ground and helping businesses make sense of their data so they can streamline their operations. Whereas a sales ceiling exists with Gotham -- i.e., Palantir won't offer access to its platform to certain countries (e.g., China) -- Foundry has a theoretical growth runway that stretches decades into the future.
Though it could take some time for Palantir to grow into its current valuation, I don't anticipate Jaluria's low-water price target being hit.
