Cathie Wood is the founder of Ark Investment Management, which operates several exchange-traded funds (ETFs) focused on innovative technology stocks. Wood thinks software companies will be the next big opportunity in the artificial intelligence (AI) industry, predicting they could generate $8 in revenue for every $1 they spend on chips from suppliers like Nvidia.
Since making that call last year, Wood has backed up her words with actions by investing heavily in leading AI start-ups like OpenAI, Anthropic, and xAI through the Ark Venture Fund. Plus, Ark's ETFs hold a number of popular AI stocks like Tesla, Alphabet, and UiPath.
If Wood proves to be right about AI software companies, here's why Amazon (AMZN -1.33%) could be among the biggest winners.
A leader in AI infrastructure and software
Amazon is one of five American companies worth $2 trillion or more, and each of them is battling for AI supremacy. But this company has one distinct advantage: It's home to Amazon Web Services (AWS), which is the world's largest cloud computing platform. It's where businesses can access hundreds of services to power their digital transition, but it's also a go-to destination for a growing suite of AI services.
AWS wants to dominate all three layers of AI: data center infrastructure, large language models (LLMs), and software. Amazon operates data centers filled with graphics processors (GPUs) from suppliers like Nvidia, but it also designed its own chips called Trainium (for AI training) and Inferentia (for AI inference). Amazon says developers can save 50% on training costs by using Trainium1 compared to competing chips, and with Trainium2 now rolling out, those savings could grow even further.
Then there is Amazon Bedrock, where developers can access industry-leading LLMs like Anthropic's Claude 3.5, Meta Platforms' Llama 3.2, and Amazon's own Titan family of models. Most developers will opt for a third-party LLM to build their AI software because it's cheaper than creating an LLM from scratch, which requires significant financial resources.
At the software layer, Amazon developed a virtual assistant called Q. It can answer questions about an organization's internal data and policies, or it can be prompted to generate computer code to accelerate software development. Amazon says Q has the highest code acceptance rate in the industry, and on one specific internal project alone, it saved Amazon $260 million and a whopping 4,500 developer years.
During the recent third quarter of 2024 (ended Sept. 30), AWS generated a record $27.4 billion in revenue. It represented a 19.1% increase from the year-ago period, which marked the fastest growth rate in 2024 so far, thanks partly to AI. Amazon says the AI business within AWS is currently growing at a triple-digit percentage rate compared to the year-ago period, and it's growing 3 times faster than the cloud division did at the same stage of its evolution.
AI is also transforming e-commerce and advertising
Although most investors are focused on AWS, e-commerce is still Amazon's largest segment, accounting for $61.4 billion of the company's $158.9 billion in total revenue during Q3. Management is trying to improve efficiency in the e-commerce segment to boost profit margins, and AI is a key part of that strategy.
For example, Amazon launched Project Private Investigator in its fulfillment centers earlier this year, which uses AI and computer vision to identify defective products before they are shipped to customers, in order to reduce the rate of returns. The company also continues to improve its Rufus virtual shopping assistant, which can answer customers' questions and even help them compare products to drive more sales.
Sellers can tap into Amelia, a powerful virtual assistant that is capable of answering operational questions, managing inventory, and even forecasting sales. Amelia can reduce costs by streamlining workflows and making sellers more efficient, which could translate into savings for customers.
Amazon also has a booming advertising business. Amazon.com receives 3.4 billion visits per month, so it's the perfect place for merchants to market their products, but the company also shows ads on platforms like Prime Video and Twitch. Businesses can access a suite of AI tools to generate text, images, and videos to instantly create engaging ads capable of converting views into sales.
Advertising revenue came in at $14.3 billion during Q3, but that figure is likely to grow significantly in the future because ad rollouts on platforms like Prime Video are still in the very early stages. Plus, since AI can turn almost any business operator into a marketing expert, Amazon's tools could help the company attract a greater share of ad spending over time.
Image source: Amazon.
Amazon stock could deliver significant long-term upside
The accelerating growth in AWS combined with the focus on efficiency in the e-commerce segment have driven a surge in Amazon's bottom line. The company generated $4.67 in earnings per share over the last four quarters, which was a whopping 143% increase from the prior four-quarter period.
Based on Amazon's stock price of $208.91 as of this writing, it currently trades at a price-to-earnings (P/E) ratio of 44.7. That might look expensive at face value, because it's a premium to the Nasdaq-100 technology index, which trades at a P/E ratio of 33.1.
However, Wall Street's consensus forecast for next year (according to Yahoo!) suggests Amazon's earnings per share could soar to $6.15. That places its forward P/E ratio at 34, which is almost in line with the Nasdaq-100:
AMZN PE Ratio data by YCharts
However, I would argue the stock deserves to trade at a premium to the Nasdaq-100. If it maintains its current P/E ratio of 44.7, that implies it could deliver a return of 31% next year. Plus, considering the pace of its earnings growth, I wouldn't be surprised if Wall Street's forecast for 2025 proves to be way too low.
As I mentioned at the top, Cathie Wood believes AI software companies could eventually generate $8 in revenue for every $1 spent on chips. Well, Amazon's capital expenditures -- most of which are going toward AI infrastructure and chips -- are on track to hit $75 billion this year, so the long-term payoff could be enormous if Wood is right.
