The Nasdaq Composite has risen steadily for nearly three years, and many believe the catalyst that sparked the current bull market was the advent of artificial intelligence (AI). Add to that the ongoing campaign of interest rate cuts and higher corporate earnings, and conditions are ripe for the market's momentum to continue. Furthermore, the tech-centric index's three-year rally suggests there will be more to come in the new year.
Going back 50 years, there have been five bull markets that have lasted longer than three years, and in each case, the rally has continued, according to Ryan Detrick, chief market strategist at financial services company Carson Group. The data shows that bull markets that persisted longer than three years continued to gain ground, lasting eight years on average. Even the shortest lasted for five years, which suggests there could be more to come.
Additionally, estimates regarding the impact of AI continue to ratchet higher. The adoption of generative AI could add as much as $15.7 trillion to the global economy by 2030, according to Big Four accounting firm PricewaterhouseCoopers (PwC), creating a windfall for those at the cutting edge of the technology.
Here are my top 10 AI stocks to buy before the Nasdaq climbs to new heights in 2026.
Image source: Nvidia.
Three chipmakers
The rise of AI was ushered in by advanced semiconductors that processed data at lightning speeds, taking sophisticated algorithms to the next level. The first to capitalize on the opportunity was Nvidia (NVDA +0.03%), positioning its graphics processing units (GPUs) as the gold standard for data center operations and ferrying data around the ether. The company's AI strategy was prescient, as it now holds an incredible 92% share of the data center GPU market. With its annual new product release cadence, rivals simply haven't been able to catch up.
If there is one downside, it's the significant energy consumption required to fuel GPUs, and the world's biggest users are beginning to take note. That's where Broadcom (AVGO 1.84%) comes in. The company developed application-specific integrated circuits (ASICs) that are viewed as a cost-effective alternative to GPUs. These specialized chips -- which Broadcom calls XPUs -- can be customized for specific tasks, making them more energy-efficient in the process.
Finally, there's Arm Holdings (ARM 3.71%). While it may not have the name recognition of Nvidia or Broadcom, Arm developed many of the blueprints and chip designs behind today's most advanced processors, licensing the CPU cores to some of the industry's biggest names -- including Nvidia. The company recently revealed that its data center customers have increased to 70,000 -- a 14-fold increase -- since 2021.
In the most recent quarter, Nvidia grew revenue 56% year over year, fueling earnings-per-share (EPS) growth of 61%. Arm sales jumped 34%, driving EPS up 120%. Broadcom grew revenue by 22% and swung from a $0.40 loss to a profit of $0.85. These results help illustrate the impact of the ongoing adoption of AI.
The most commonly used valuation metrics struggle to value high-growth companies, and our trio is no exception. However, when measured using the forward price/earnings-to-growth (PEG) ratio, Nvidia, Arm, and Broadcom have multiples of 0.76, 0.73, and 0.06, when any number less than 1 suggests an undervalued stock.
NASDAQ: NVDA
Key Data Points
One chip foundry
We've established that it takes highly specialized semiconductors to support the AI revolution -- and Taiwan Semiconductor Manufacturing (TSM 0.95%), commonly called TSMC, is the foundry to the AI stars. In fact, Nvidia, Arm, and Broadcom are all customers. There's more. TSMC developed one of the most intensive and complicated chipmaking processes, parlaying that into unmatched economies of scale.
The company fabricates an estimated 90% of the world's most advanced semiconductors, essentially cornering the market. Third-quarter revenue grew 30% year over year, while diluted EPS jumped 39%. Despite TSMC's critical position in the AI ecosystem, the stock is currently trading for 23 times next year's expected earnings.
Four cloud providers and one social media titan
Developing the large language models (LLMs) that underpin AI is a costly affair, and the amount of data required is enormous. As such, there aren't many companies with the resources to develop these systems. The biggest cloud infrastructure providers are among the select few with the combination of money and data necessary to bring these AI models to life.
As the four largest cloud providers, Amazon (AMZN +0.64%) Web Services (AWS), Microsoft (MSFT 0.04%) Azure, Alphabet's (GOOGL 2.08%) (GOOG 1.94%) Google Cloud, and Oracle (ORCL 1.86%) Cloud Infrastructure (OCI) have a strategic advantage when it comes to AI. Not only do they have the resources necessary to create AI models, but each also has a user base that acts as both captive audience and target market for their AI services. Furthermore, each company is using its models in-house to streamline processes and generate efficiency gains (read "increase profits").
Finally, each of these companies generates significant revenue from a primary business: Amazon is the global e-commerce leader and a leading provider of digital advertising; Microsoft is a leading enterprise software and software-as-a-service (SaaS) provider; Alphabet is the global search and digital advertising leader; Oracle is a leading provider of information technology (IT) services.
Image source: Getty Images.
While Meta Platforms (META +0.50%) isn't a cloud provider, it does have the data and resources to create its own LLMs, thanks to 3.5 billion daily visitors to its various social media platforms and the resulting advertising revenue. The company used its user data to create the first large-scale open-source AI models, which it now offers to the cloud providers -- for a fee, of course.
Our quintet of AI players generated robust results overall in the most recent quarter:
- Amazon's revenue grew 13% year over year, while EPS grew 36%.
- Microsoft's revenue climbed 18%, fueling adjusted EPS that grew 23%.
- Alphabet's revenue rose 16%, while EPS jumped 35%.
- Oracle's revenue increased 12%, pushing adjusted EPS up 6%.
- Meta's revenue jumped 26%, driving adjusted EPS up 20% (excluding one-time charges).
Finally, each of these five stocks is reasonably priced relative to the opportunity. Amazon, Oracle, and Microsoft are selling for 32 times, 31 times, and 27 times next year's expected earnings, respectively. At the same time, Alphabet and Meta are even more attractively priced, selling for 26 times and 21 times next year's earnings, respectively.
NASDAQ: PLTR
Key Data Points
One viral AI pioneer
Perhaps more than any other player, Palantir Technologies (PLTR +1.65%) has leveraged its decades of experience to develop AI solutions that provide actionable insights in real time. The company's Artificial Intelligence Platform (AIP) has become the belle of the AI ball, taking enterprise software and government users by storm.
Recent results paint a rosy picture. In Q3, Palantir's U.S. commercial segment -- which includes AIP -- accelerated 121% year over year and 29% sequentially, marking the ninth consecutive quarter of acceleration. This helped fuel a 199% increase in its remaining deal value (similar to backlog).
This drove the segment's total contract value (TCV) up 342%. Not only is the company signing more agreements, but the value of those deals is also steadily increasing: 204 were worth at least $1 million, 91 worth at least $5 million, and 53 worth at least $10 million. Furthermore, its Rule of 40 score of 114% is off the charts.
That said, Palantir won't appeal to every investor. The stock is currently selling for 179 times next year's expected earnings, which is pricey by any measure. However, as my colleague Harsh Chauhan pointed out, its performance could make Palantir the best-performing AI stock of 2026.
