Artificial intelligence (AI) has been referred to as a generational opportunity by leadership at some of the top tech companies in the world. While some of the benefits are already beginning to be felt, AI still looks to be in the early stages of its development as a revolutionary technology.
Given that there is still a long way to go in AI's development, there is also seemingly plenty of opportunity left for companies to benefit and potentially help make investors who get in early into millionaires. Let's look at three stocks across AI infrastructure, cloud computing, and software that could push a millionaire-making portfolio to its goal over the long haul.
1. Nvidia
Nvidia (NVDA -1.96%) has been one of the biggest AI winners thus far, but that doesn't preclude it from continuing to be a big winner. Computing power is paramount for training large language models (LLMs) and running inference needed for AI's development and operations. Nvidia's graphics processing units (GPUs) have proven to be one of the best chips to accomplish these tasks. One of two big GPU designers, Nvidia has been able to take a dominant market share of nearly 90% in the space with the help of its CUDA software platform, which makes it much easier to program GPUs for various AI tasks.
As such, Nvidia's GPUs have become the backbone of AI data center infrastructure, which continues to grow at a rapid pace. As companies look to train more sophisticated LLMs, they need exponentially more computing power. Some of the latest versions of AI models require 10 times as many GPUs as their prior versions in order to accomplish their computationally intensive tasks. There are indications that there could be LLMs requiring 1 million AI chip clusters in the near future.
Meanwhile, AI infrastructure remains robust, with Nvidia's largest customer Microsoft recently announcing it would spend an astonishing $80 billion building out data centers in 2025. Other large tech companies have also indicated that they plan to increase their AI-related spending this year.
With no end currently in sight for AI infrastructure spending, Nvidia continues to look like a long-term winner. Meanwhile, the stock is attractively priced with a forward price-to-earnings (P/E) ratio of below 30 and a price/earnings-to-growth ratio (PEG) under 1, which generally indicates a stock is undervalued.
2. Alphabet
Alphabet (GOOGL -1.35%) (GOOG -1.30%) is hoping to be an AI winner in several areas. The company has already seen strong growth in its cloud computing unit, Google Cloud, which saw its revenue jump 35% last quarter, as it helps customers build out their own AI models and applications. Meanwhile, the high-fixed-cost business has grown to a point where it is now profitable. As a result, segment income surged from $266 million in the year-ago quarter to $1.95 billion in the third quarter of 2024.
Within the data center segment of its business, Alphabet has also designed its own custom AI chips with the help of Broadcom. By using these custom tensor processing units (TPUs) in combination with GPUs, Alphabet is improving AI inference processing times and lowering costs for its users and clients. That could help give the segment even more operating leverage.
In addition, Alphabet has a nice opportunity with AI search and its Gemini AI assistant app. The company historically has only served ads on about 20% of its searches, so eventually monetizing its AI Overviews and Gemini app results would be a nice future opportunity. The company has also taken the lead in multimodal AI, with things such as text-to-video generation and visual search.
The company is also a leader in some other AI-adjacent emerging technologies including quantum computing and autonomous driving. The stock, meanwhile, is attractively valued, trading at a forward P/E of 19 times.
Image source: Getty Images
3. Salesforce
While AI infrastructure and cloud computing companies have seen the biggest revenue growth, agentic AI could be the next big wave. The AI getting the most attention at the moment has been generative AI, where AI can produce images, sound, or text in response to a user query. Agentic AI takes that a step further -- AI agents can go out and autonomously perform tasks within given parameters with little human supervision.
Salesforce (CRM -0.36%) is looking to be at the forefront of agentic AI with its new Agentforce solution. The company already has a large customer base with its customer relationship management (CRM) software platform, and it will look to upsell these customers to its new AI agents. The company will have a number of out-of-the-box agents that can perform tasks in such areas as customer service, sales and marketing, and recruiting, while they will also be customizable through its no-code and low-code tools.
Released only in October, Agentforce has grown quickly, with the company saying it had more than 1,000 deals in place in mid-December. It projects to have 1 billion agents in force in fiscal 2026 (ending January 2026). Agentforce is a consumption product that costs $2 per conversation, so the more useful customers find its agents, the more upside it has.
Salesforce currently trades at a forward price-to-sales (P/S) ratio of 7.4 times fiscal 2026 analyst estimates, which is very reasonable given the opportunity in front of it.
