The artificial intelligence (AI) arms race is a heated contest. On one side lies Big Tech. In particular, Microsoft and Alphabet have invested billions into large language models (LLMs) and other AI-powered applications. On the other side are integrators such as Cathie Wood favorites UiPath and Twilio or data services platforms such as Snowflake.

But in the midst of all of the AI hype, one Big Tech firm seems to be less dominant in the headlines. During its second-quarter earnings call, the chief executive office of e-commerce and cloud computing leader Amazon (AMZN 1.12%) provided in-depth details about the company's AI ambitions. However, given the competition's splashy investments in OpenAI and other unicorns, I feel that these comments went largely unnoticed. 

Since reporting Q2 results, Amazon has made a number of strategic chess moves as it builds out its AI roadmap. I think now could be a really interesting opportunity to pounce on some shares of this Warren Buffett stock before AI-driven momentum potentially enters the picture.  

Artificial intelligence has many layers

On Amazon's second-quarter earnings call, CEO Andy Jassy provided investors with an in-depth look into how Amazon thinks about artificial intelligence. 

The first layer that Jassy speaks about is the computational power needed to train LLMs to make better inferences or predictions. Then, in the middle, Amazon believes that it can market AI as a managed service. What this means is that businesses could rely on the technical expertise of other enterprises, such as Amazon, which can be the source of generative AI models so that corporations do not need to spend the time and capital building LLMs in-house. Jassy calls the combination of the first two layers a means of "democratizing access to generative AI." The top layer, in Jassy's view, is the final product that actually gets deployed. He cited ChatGPT as a specific application that resides on this third layer. 

Although this is a nice summary of the different puzzle pieces that come together to build a broader AI picture, what exactly is Amazon doing to capitalize on these trends?

A person using a chatbot to answer questions.

Image source: Getty Images.

What is Amazon doing to compete?

As referenced above, the first layer revolves around computational power. During his commentary, Jassy pointed out that Amazon has been developing its own semiconductor chips for several years. More specifically, the company is building models for training and inferencing called Trainium and Inferentia, respectively. This could prove to be a lucrative catalyst for Amazon as the company seeks to gain market share from the likes of AMD and Nvidia. While Amazon's chip endeavors are in the early days, the company just struck a huge partnership with AI start-up Anthropic, which will be training its models on Trainium and Inferentia.

The middle layer is where things begin to get really interesting. Amazon has developed a service called Bedrock for its AI-managed service offering. On the call, Jassy said:

Stepping back for a second, to develop these large language models, it takes billions of dollars and multiple years to develop. Most companies tell us that they don't want to consume that resource building themselves. Rather, they want access to those large language models, want to customize them with their own data without leaking their proprietary data into the general model, have all the security, privacy, and platform features in AWS work with this new enhanced model, and then have it all wrapped in a managed service.

What Jassy is really trying to convey here is that while AI is now one of the hottest topics for businesses of all sizes, very few have the capital or intellectual horsepower to build solutions without partners. Moreover, generative AI applications are not an overnight project. For this reason, enterprises such as Amazon, Alphabet, and Microsoft, all of which have been investing in artificial intelligence for many years, are likely to be heavily relied on for outsourced AI needs. While Bedrock sounds like an optimal solution, Jassy did not hesitate to showcase how popular the platform already is. For example, companies such as 3M and Ray Dalio's Bridgewater Associates are both Bedrock customers.

Amazon's move into chips and its AI-as-a-service offering are already looking to be meaningful tailwinds for the company's cloud platform, Amazon Web Services (AWS). This is even more important when investors consider that Google Cloud and Microsoft Azure have been gaining meaningful market share from AWS.

Should you invest in Amazon?

AMZN Chart

Data source: YCharts.

The charts above are simple illustrations of the stock price gains in Amazon and some of its competitors, as well as the S&P 500 over the past year. The obvious takeaway is that many of Amazon's Big Tech counterparts have witnessed significant increases, and the broader market has even outperformed Amazon.

Amazon's core e-commerce business has felt the effects of inflation and high borrowing costs. Consumer discretionary buying habits have shifted given these macroeconomic variables, and companies of all sizes have been operating under tight budgets, resorting to layoffs to trim expenses. Amazon is not immune to these challenges, and its stock price reflects investor sentiment.

With all of this said, Amazon stock might be oversold. The company's Q2 results had several highlights, including a return to positive free cash flow. When you consider that Amazon has been able to move forward and pursue some interesting new foundations as it relates to AI, the company's depressed stock price begins to look more appealing.

AMZN PE Ratio Chart

Data source: YCharts.

The stock is trading well below its past price-to-earnings (P/E) highs, which could signal that investors do not place the same premium outlook on the business as compared to other names in Big Tech. While Alphabet and Microsoft are becoming synonymous with AI, I wouldn't leave Amazon out of the discussion just yet. I think now is a great opportunity to lower your cost basis in Amazon before any meaningful benefit from the Anthropic deal or its new cloud tailwinds occur.