From Search Boxes to Agentic Commerce
AI shopping assistants are conversational systems embedded into retail sites and apps that move beyond search boxes by acting as active agents that interpret intent, ask follow-up questions, suggest options, and personalize recommendations across the entire shopping journey, so customers spend less time browsing and more time receiving context-aware guidance tailored to their tastes, constraints, and goals. This shift marks a new phase of retail AI technology often called agentic commerce, where software agents do more than retrieve products; they shape decisions. Instead of typing keywords and scrolling filters, shoppers describe occasions, budgets, or styles in natural language and the AI narrows choices. For retailers, these agents promise higher conversion and loyalty by turning static product catalogs into guided experiences that adapt in real time to each customer’s behavior and feedback.
Amazon Turns Alexa for Shopping into a Retail AI Platform
Amazon Web Services is pushing agentic commerce into the mainstream with its new AWS Agentic Shopping Assistant, built on the same technology behind the Alexa for Shopping assistant on Amazon.com. Amazon says Alexa for Shopping, formerly known as Rufus, drove nearly USD 12 billion (approx. RM55.2 billion) in incremental sales last year, and it now wants outside retailers to tap that engine. The assistant can converse with shoppers, answer product questions, and tailor recommendations to each store’s inventory, branding, and business rules, with AWS claiming retailers can launch in about 60 days. Amazon’s strategy is clear: sell retail AI technology as a service while keeping customer data and catalogs under each retailer’s control. Early adopters like Kate Spade are already experimenting with gift concierge experiences that ask about the occasion and recipient before proposing products, turning browsing into guided decision-making.
Kmart’s Joy Blends Virtual Try-On with Conversational Advice
Kmart is pairing AI shopping assistants with virtual try-on technology to create a more tactile online experience. Its new assistant, Joy, will live on the website and app, helping shoppers discover products, compare options, and refine searches using natural language prompts. Customers can specify size, style, color, and budget and receive curated suggestions within seconds, or upload photos to get tailored recommendations across Kmart’s marketplace range. Joy is tightly integrated with features like “See It in My Space,” which lets people visualize items and furniture in their own homes before buying, and virtual try-ons for selected products. These capabilities are powered by Gemini Enterprise for Customer Experience from Google Cloud, an agentic solution that can support customers from discovery to post-purchase resolution. As Kmart’s customer chief puts it, “Customers aren’t just searching anymore; they’re engaging conversationally and looking for ideas and guidance.”
Inside Shopify: Agentic AI Workflows for Merchants, Not Shoppers
While Amazon and Kmart focus on shopper-facing AI customer experience, Nosto’s latest work with Shopify Sidekick targets the teams behind the storefront. Nosto, an agentic commerce experience platform, has integrated its product discovery and personalization tools directly into Shopify’s AI assistant so merchandisers can adjust recommendation layouts and logic through conversational instructions. In a live demonstration with Ford’s ecommerce partner Autonative, prompts given to Sidekick modified the design of Nosto-powered product recommendations on an accessories store without leaving Shopify or calling developers. The goal is to remove the operational bottlenecks that slow down testing and optimization of AI shopping assistants on the front end. Nosto’s roadmap goes further, with its AI agent Huginn connecting through MCP and the Huginn Connect technology to power agentic workflows such as technical guidance, proactive merchandising insights, and, eventually, automated execution of recommended actions.
What This Wave of AI Shopping Assistants Means for Customers
Together, these projects show retail AI technology evolving toward end-to-end agentic commerce. Amazon’s AWS Agentic Shopping Assistant offers a powerful, white-label engine for conversational discovery at scale. Kmart’s Joy emphasizes visual immersion through virtual try-on technology and “See It in My Space,” making digital shopping feel closer to in-store exploration. Nosto’s work inside Shopify Sidekick focuses on empowering teams to build and refine these AI customer experiences without heavy technical overhead. For shoppers, the common thread is fewer menus and filters and more dialog-driven guidance that fits personal needs. For retailers, the stakes are sizeable: Accenture estimates that by 2030 more than 30% of online commerce could run through AI agents, representing about USD 3.1 trillion (approx. RM14.26 trillion) in transactions. The race now is to build assistants that are not only smart, but trusted and transparently aligned with each brand.