Amazon’s new AI cloud strategy is taken straight from the e-commerce playbook that built a $2 trillion behemoth

These have been critical keys to Amazon’s e-commerce dominance over the past two decades.

Now the tech giant appears to be doubling down on that winning formula, borrowing parts of it to drive a new AI strategy from its Amazon Web Services division, focused in part on low prices and wide selection.

To be clear, it will be a long time before the business community can assess the effectiveness and financial sustainability of Amazon’s approach. But if successful, the plan would go a long way toward both appeasing critics who argue that Amazon is playing catch-up in the AI ​​wars and future-proofing the company’s reputation as one of the world’s most powerful and influential tech conglomerates for decades to come.

Amazon executives unveiled key parts of their AI strategy at their flagship AWS Re:Invent conference in Las Vegas this week. A key element is a new portfolio of home-developed base models, so-called FMs or LLMs, called Nova, which can process text, image and video queries.

The introduction of a new class of Amazon’s own AI models might be confusing at first glance, as the company has already invested $8 billion in Anthropic, the maker of the popular Claude family of genetic AI models. But as my colleague Sharon Goldman recently noted, Amazon believes there will “never be a single tool” – or AI model – to rule them all. In short, Amazon assumes that companies want a choice of models, be it from Amazon, Anthropic or other tech giants like Meta.

Amazon actually pushed this idea of ​​offering business customers a choice of AI models through a single API when its AWS division launched a service called Amazon Bedrock last year. Through Bedrock, business customers could choose from a relatively limited selection of AI models—but a selection nonetheless—to train for their own needs and to serve as the basis for their own genetic AI applications.

On Wednesday, Amazon doubled down on its strategy and announced the Bedrock Marketplace, offering a total of 100 AI models. The LLMs on the market come from a variety of different companies, some of which are designed for specific purposes.

“Finding and evaluating these models can be difficult and costly,” Amazon said in its blog post announcing the marketplace. “You have to discover them across different services, create abstractions to use them in your applications, and create complex security and governance layers. Amazon Bedrock Marketplace addresses these challenges by providing a single interface for access to both specialized and general (baseline) models.”

Looking back at Amazon’s e-commerce business, the Amazon Marketplace is a core component where hundreds of thousands of third-party merchants offer products for sale, accounting for 60% of all goods sold on Amazon. Amazon supplements the offering of these marketplace sales by selling its own inventory, sometimes under its own brand name, if a particular product category or price range is not covered by the marketplace sellers or Amazon’s brand partners.

Similarly, Amazon offers businesses an enterprise AI version of the marketplace, which one can imagine will only expand the choice in the future. (It’s also worth noting that Amazon’s core business, AWS, offers a marketplace of more than 10,000 software tools, covering categories from cybersecurity to data analytics.)

Low prices have also been another hallmark of Amazon’s retail business. Amazon is aggressively matching other retailers’ prices and running two huge discount promotions that are attracting big spenders and new Prime customers. (The FTC has argued in its antitrust lawsuit against Amazon that the e-commerce giant artificially inflates consumer prices online by penalizing merchants who sell products at a lower price at other retailers, but that’s an issue for another day.)

And sure enough, Amazon CEO Andy Jassy’s presentation began with price: “75% cheaper,” was the first feature.

Simon Willison, an independent AI researcher, conducted a quick test and agreed, writing on the social network app Bluesky that Amazon is “competitive in price and performance with the Google Gemini family, which means it are _really_ cheap.”

“With this release, I believe Amazon has secured a place in the top echelon of model providers,” Willison added. “Maybe we need a new FAANG acronym that covers OpenAI, Anthropic, Google, Meta and Amazon. I like GAMOA.”

Amazon executives don’t care what the new acronym is called, as long as they earn a place in it. If this is the case, the company’s long-standing hallmarks of low prices and selection will likely be the main reasons behind it.

Are you a current or former Amazon or AWS employee with thoughts on this topic or would you like to share a tip? Contact Jason Del Rey at [email protected], [email protected]or via the messaging apps Signal and WhatsApp at 917-655-4267. You can also send him a message on LinkedIn or at @delrey To X.

How far are you from the world’s most powerful business leaders? Discover who made our brand new list of the 100 most powerful people in business. Plus, learn more about the metrics we used to create it.