Swiggy has rolled out AI-powered ordering across its food delivery, quick commerce and dining-out services, allowing users to place orders directly through conversational AI platforms such as ChatGPT, Claude and Google Gemini.

The integration is powered by Model Context Protocol (MCP), an open-source framework that allows AI assistants to connect with live external services. With this setup, users can browse menus and products, add items to their cart, apply offers, complete checkout and track orders-all within a chat interface, without opening the Swiggy app.

The rollout spans Swiggy Food, Instamart and Dineout. Swiggy stated that Instamart is the first quick-commerce platform to adopt MCP, offering access to more than 40,000 products through natural language prompts. Users can issue simple commands such as ordering groceries for a specific recipe, choosing restaurants based on ratings or cuisine preferences, and booking tables at selected venues.

Once a request is made, the AI assistant searches available options, compares prices, applies eligible discounts and completes the transaction within the conversation. For dining reservations, the system can check availability, retrieve time slots and confirm bookings in a single interaction. Post-checkout, users can also receive real-time order updates and track deliveries through the same interface.

According to Swiggy, the integration is aimed at simplifying the ordering journey by reducing screen navigation and enabling intent-based interactions. Users can connect Swiggy services to supported AI assistants by adding a custom connector and selecting the appropriate MCP endpoint for food delivery, grocery shopping or dining reservations.

With this move, Swiggy positions itself at the intersection of commerce and conversational AI, expanding how users interact with on-demand services.