AI Shopping Bots Race to Fill Online Carts

Silicon Valley is unleashing artificial intelligence agents that can surf online stores, swipe digital credit cards and buy goods without a human lifting a finger — forcing eCommerce players to grapple with a future where the shoppers aren’t people.

Tech heavyweights and startups are racing to crack the code of automated buying, a development that could reshape how retail platforms handle everything from price competition to checkout flows. For example, Perplexity recently launched an AI agent shopping tool last month that lets U.S. subscribers buy through its platform. It offers one-click checkout for select merchants and visual product search capabilities. The company also opened a merchant program providing retailers with product indexing and checkout integration.

“For commodity products, AI shopping agents will intensify price competition, potentially driving a ‘race to the bottom,’” Nigel Daley, chief operating officer and co-founder of AI search company Vantage Discovery, told PYMNTS. “With the ability to identify the lowest price in seconds, these agents will make opaque pricing strategies obsolete. To stay competitive, retailers must adopt transparent pricing, implement dynamic pricing to adjust based on market conditions, or introduce value-driven offerings, such as personalized discounts/bundles, to stand out without solely competing on price.”

Interest is growing in AI shopping agents. Amazon rolled out its AI-powered shopping assistant, Rufus, to all U.S. customers, making personalized shopping easier. Google unveiled Gemini 2.0, an advanced AI model designed to act like a virtual personal assistant capable of handling tasks and interacting naturally with users. Meanwhile, HotelPlanner introduced AI travel agents capable of handling end-to-end hotel bookings in multiple languages. The AI system processes approximately 10,000 customer calls per day.

AI shopping agents can be everything from simple chatbots that give personalized suggestions based on browsing data to all-out platforms, where these tools aggregate data from across the web to help users find the best prices on the products they want, Buysmart.AI founder Lifei Chen told PYMNTS.

“Often, they’ll have extra value-added features, like pricing history and current prices across eCommerce sites with discounts factored into the price,” he said. “Rather than having to price compare, AI can do all the leg work for you.”

Early versions of AI agents are already changing shopping, offering tailored recommendations and streamlining tasks. Amazon’s AI assistant, Rufus, provides personalized product suggestions and answers customer questions. Microsoft’s Copilot Vision integrates with the Edge browser, suggesting products and reading handwritten notes like recipes. Encore, an AI search tool, helps users find vintage and secondhand items across resale websites.

“AI shopping agents have the potential to find products or deals that you might not otherwise have found on your own, across more retailers/brands than a person would normally be able to find on their own,” ReverseLogix founder and CEO Gaurav Saran told PYMNTS. “In theory, these tools could save consumers hours while also helping you easily locate products or brands you didn’t even know existed. This could level the playing field in some product categories or, at the very least, allow smaller merchants to be found in the same search results as Amazon.”

While AI promises to handle the heavy lifting of comparison shopping, there’s one weight these digital assistants aren’t ready to shoulder: your wallet. Security remains the elephant in the virtual shopping aisle, according to Saran, who pointed to payment automation as the critical challenge ahead.

“If you only innovate the product search but do not optimize the purchase process, you are leaving money on the table or money that can be spent elsewhere with another merchant we can fulfill on the entire journey,” he said.

Creating complex dynamic pricing algorithms will be essential to make AI shopping assistants as quick and reliable as real people, Michelle Nguyen, product owner and marketing manager at UpPromote, an affiliate marketing company, told PYMNTS.

“With these algorithms, the agents could change prices right away based on things like competition in the market, customer demand and the amount of stock on hand,” she said. “In this way, the agents would be able to choose purchases that get the best deals, making sure that customers get the best prices.”

For all PYMNTS AI coverage, subscribe to the daily AI Newsletter.

The post AI Shopping Bots Race to Fill Online Carts appeared first on PYMNTS.com.