Published Date : 8/10/2025
The largest tech companies are investing heavily in data centers to meet the growing demand for artificial intelligence (AI). These massive expenditures are justified by the transformative potential of AI, which is expected to drive significant economic growth. For instance, Morgan Stanley estimates that long-term efficiencies from AI could be worth $40 trillion.
If this estimate is accurate, investors are not bullish enough on AI. The potential for unprecedented economic growth over the next few decades is substantial. Buying and holding the stocks of top AI companies could help you realize significant gains. Here are two top AI stocks to consider for long-term investment.
1. Palantir Technologies
The growing adoption of AI is playing to the strengths of Palantir Technologies' differentiated technology. Its AI platform deeply integrates with a company's operations, intelligently turning data into actionable insights. The company started as a defense contractor but has successfully expanded into the commercial market.
For example, Citibank has reduced the time to onboard customers from days to just seconds, and Fannie Mae is using Palantir to detect mortgage fraud almost instantly, down from two months. These are just a few examples of the significant efficiency gains companies are experiencing by using Palantir.
Palantir also continues to do a lot of business with the U.S. government, recently signing a 10-year agreement with the U.S. Army worth up to $10 billion. This is nearly triple Palantir's trailing-12-month revenue of $3.44 billion. In the second quarter, Palantir posted a year-over-year increase in revenue of 48%, up from a 39% growth rate in the previous quarter. Management expects revenue to accelerate again to 50% in Q3.
Palantir's differentiated offering and accelerating growth are why the stock is trading at a high valuation. While no software company has ever sustained a price-to-sales multiple of 136, Palantir's financial results are impressive. For example, while small, fast-growing software companies usually report losses in their early growth phase, Palantir is already earning a stellar profit margin of 33% in Q2, on par with Microsoft.
Its margin should continue to increase as the business scales. CEO Alex Karp believes the business can increase revenue by 10x with fewer employees. At the rate it is growing, it can probably achieve that within seven years. Wedbush Securities analyst Dan Ives believes Palantir will reach a $1 trillion market cap by 2028, more than double its current value.
2. Nvidia
Nvidia has been the poster child of the AI boom over the last few years, and it's likely to maintain this position. Every data center must have Nvidia chips to be competitive. The company offers a complete stack of solutions across processors, networking, and software, giving Nvidia a competitive advantage.
All the leading cloud service providers and AI model builders, including ChatGPT maker OpenAI, are using Nvidia's Blackwell chips for training AI in large data centers. Selling these sophisticated computing products is Nvidia's main growth driver. In the last quarter, its data center revenue grew 56% year over year.
Nvidia is building a strong competitive advantage around its data center chips. This is evident in the robust growth of its networking solutions, which made up 15% of its revenue last quarter but grew an impressive 98% year over year. Earlier this year, Nvidia launched NVLink Fusion, which can integrate custom chips from other semiconductor companies into its computing systems and software. This further stretches its addressable market and protects its competitive lead.
Tying all this together is Nvidia's CUDA software, which has seen the number of developers double over the last three years to 5.9 million. CUDA allows customers to optimize Nvidia's GPUs for a wide variety of applications. The network of developers using CUDA has been fundamental to Nvidia's leadership in the GPU market for the last 20 years.
Despite the stock's amazing run, it is still a solid investment that could hit new highs over the next year. This is clear by looking at the stock's forward price-to-earnings multiple on next year's earnings estimates, which is currently 30. This is a very conservative valuation for almost any high-growth business and suggests that Nvidia is still undervalued.
Q: What is the estimated long-term economic impact of AI according to Morgan Stanley?
A: Morgan Stanley estimates that the long-term efficiencies from AI could be worth $40 trillion.
Q: What significant agreement did Palantir Technologies recently sign with the U.S. Army?
A: Palantir Technologies recently signed a 10-year agreement with the U.S. Army worth up to $10 billion.
Q: What is the main growth driver for Nvidia's data center revenue?
A: The main growth driver for Nvidia's data center revenue is the sale of its Blackwell chips for training AI in large data centers.
Q: How many developers are currently using Nvidia's CUDA software?
A: Nvidia's CUDA software is currently used by 5.9 million developers.
Q: What is Nvidia's forward price-to-earnings multiple on next year's earnings estimates?
A: Nvidia's forward price-to-earnings multiple on next year's earnings estimates is currently 30.