Amazon's AI Ambitions: From Shopping Guides to Autonomous Agents

Amazon Wants AI to Do Your Shopping for You

Amazon, building on some AI systems it's already "sprinkled" through its e-commerce platform, has announced plans for AI-powered shopping guides for "hundreds of different product categories," tech site Wired reports. This is a classic use of current-gen AI tech, which is great at summarizing lots of information down to bite-size chunks -- useful for a shopper who may be blinded by thousands of different options when looking for one particular product. Amazon executives also told Wired that they have much more ambitious plans for AI, centering on AI agents. Agents are small pieces of software built on core AI technology that can be programmed to carry out particular tasks: OpenAI's CEO Sam Altman has heralded them as the future of AI, and it's easy to see that if an agent can automate humdrum office tasks for you, then they could save many people time and money. In its look at Amazon's strategy, Wired explains that agents may start off performing relatively simple tasks: like spotting when an author releases a new book in a series you've been reading, and then automatically adding it to your shopping cart -- and possibly even authorizing the purchase, if you've given it permission. Rajiv Mehta, a vice president at Amazon working on conversational AI, told Wired that Amazon has been looking at how ads can be built into this process. Because Amazon is as much an ad platform as an online retailer.

Amazon Wants AI to Decide What You Buy

The tentacles of AI hype have already wriggled their way into Hollywood studios, record labels, newsrooms, kitchens, and maybe even the caverns of your mind. Amazon believes your shopping cart is next. The retail giant is apparently laying the groundwork for AI-initiated purchases, starting with automatically generated buying guides; they're out today for around 100 kinds of products, including TVs, dog food, shoes, and face creams. Amazon says the guides already point out specific features, use cases, brands, and answers to common questions. They're currently exclusive to smartphones, and they sound a whole lot like an automated alternative to Wirecutter, but without the affiliate links and the sense that a human actually waded through the muck to find the best thing. Amazon's intention with these AI-generated guides, according to the company's head of personalization Daniel Lloyd, is to "reduce the time spent researching before you make a purchase." Amazon really wants to chip away at this, though Lloyd acknowledged that "it's still early days." The new guides, along with Amazon's shopping chatbot, Rufus, are precursors to more sophisticated shopping agents that could simply plop items directly into your cart and effectively click buy, Amazon AI executive Trishul Chilimbi told Wired. More sophisticated agents are part of the company's "roadmap," Chilimbi reportedly said, informed by the heaps of shopping data Amazon already collects. "We’re working on it, prototyping it, and when we think it’s good enough, we’ll release it in whatever form makes sense," he added. The path to such agents may start with chatbots that proactively ping you with product recommendations and ads, according to Chilimbi, but Amazon also says it's planning for a future where you might just give it a prompt and a budget, and automate the research, browsing, and check-out process altogether. Tech giants have raced to pack AI features into their products over the past several years, and in some cases, their efforts have driven down the value of human labor while at least momentarily supercharging valuations. The AI bubble has also spawned plenty of as-of-yet unfulfilled promises, including the loosely defined specter of artificial general intelligence. While the AI bubble may actually be deflating, Amazon certainly knows how to wield a tech prediction to boost sales, as it did with drone delivery hype during the 2013 holiday shopping season. Now, Amazon seems to have done it again, drawing the eyes of skeptics and shoppers alike towards a (sigh) mountain of deals.

Amazon Dreams of AI Agents That Do the Shopping for You

Amazon might not have ChatGPT, but it has a roadmap that includes developing even more advanced forms of artificial intelligence -- including AI agents that are hell-bent on helping you buy stuff. The ecommerce company is already sprinkling ChatGPT-like AI over its website and apps -- today announcing, among other enhancements, AI-generated shopping guides for hundreds of different product categories. Executives at the company say its engineers are also exploring more ambitious AI services, including autonomous AI shopping agents that recommend goods to a customer or even add items to their cart. "It's on our roadmap. We're working on it, prototyping it, and when we think it's good enough, we'll release it in whatever form makes sense," says Trishul Chilimbi, a VP and distinguished scientist at Amazon who works on applying the company's core AI to its products and services. Chilimbi says the first step toward AI agents will likely be chatbots that proactively recommend products based on what they know of your habits and interests, as well as a grasp of broader trends. He acknowledges that making this feel nonintrusive will be crucial. "If it's no good and annoying, then you'll tune it out," he says. "But if it comes up with surprising things that are interesting, you'll use it more." Amazon added a chatbot called Rufus to its platform in February 2024 that can answer a wide range of questions about Amazon's many products. The bot uses a bespoke large language model -- the kind of algorithm that powers ChatGPT -- also known as Rufus. The Rufus LLM is trained on vast amounts of internet text -- like those on publicly available websites -- and is then fine-tuned into a commerce-focused model by being fed a carefully curated diet of Amazon's proprietary data. Chilimbi says Amazon's LLM has "hundreds of billions of parameters." (Parameters are a rough measure of capability; for comparison, Meta's largest publicly available LLM has 405 billion.) He confirmed that Amazon is training a larger model, but declined to say how large it is or what capabilities Amazon hopes it will unlock. Like many tech companies, Amazon is looking beyond chat and turning its attention toward the potential of so-called agents, which use LLMs but attempt to carry out useful tasks on users' behalf either by writing code on-the-fly, inputing text, or moving a computer's cursor. Future AI agents might, for instance, navigate various websites to sort out a parking ticket, or they might operate a PC to file a tax return. Getting LLM-powered programs to do this reliably is elusive, however, because such tasks are vastly more complex than simple queries and require a new level of precision and reliability. "Every major company is now doing [AI] agents," says Ruslan Salakhutdinov, a computer scientist at Carnegie Mellon University who is working on AI agents. The technology is exciting, he says, because it promises to automate countless routine tasks that people do every day: "On the ecommerce side, if agents can find the best possible outcome for me, that's amazing." Salakhutdinov and colleagues at CMU developed a dummy ecommerce website as part of a platform called Visual Web Arena for testing AI agents. Key challenges include enabling agents to better make sense of visual information and training them to explore vast arrays of possible options while zooming in on the correct one -- something that may require more advanced reasoning abilities.