Google is testing a 'Buy' button in Gemini and AI Mode for Flipkart products in India, shifting AI from product discovery to direct e-commerce transactions.

The boundary between AI-driven research and the final act of purchasing is blurring. Google has initiated a strategic experiment in India, allowing users to buy products from Flipkart, the Walmart-owned e-commerce giant, directly through Gemini and the AI Mode integrated into Google search. This development marks a critical pivot for the search giant: moving beyond the role of a digital concierge that recommends products toward becoming a transactional engine that closes the sale.

The mechanics of the AI Buy button

For a limited group of users in India, the shopping experience within Gemini has evolved. Instead of receiving a list of links or product descriptions that require a jump to an external app or browser tab, these users now see a prominent Buy button on select Flipkart listings. When clicked, this button triggers a checkout flow branded by Flipkart, allowing the transaction to be completed without the user ever leaving the AI interface.

This integration streamlines the consumer journey by removing the friction of app-switching. While the purchase is initiated within the AI environment, the actual checkout process is handled by Flipkart's own systems. This differs from previous Google demonstrations where the company showcased a Google-hosted checkout experience, suggesting that Google is testing multiple architectural approaches to AI commerce.

Strategic focus on high-value electronics

The current phase of the test is highly targeted. Google has not opened the floodgates to all categories; instead, the trial is restricted to a small selection of products. The primary focus remains on high-demand tech categories, specifically:

Smartphones, mobile accessories, and various electronics.

While some users experience this seamless transactional flow, others continue to see standard product listings. This A/B testing approach allows Google to measure the conversion lift provided by the direct Buy button compared to traditional referral links. The timing is not accidental; sources indicate that Google intends to expand this rollout in late October to capitalize on India's massive festive shopping season, a period characterized by peak consumer spending and aggressive marketplace competition.

Agentic commerce and the Universal Commerce Protocol

This experiment is a practical application of what Google calls agentic shopping. The goal is to create AI agents capable of navigating the entire shopping journey, from the initial query to the final payment. To support this vision, Google previously introduced the Universal Commerce Protocol (UCP).

UCP is designed as an open standard that enables AI agents to interact with various retailers. While the current Flipkart test uses a marketplace-branded flow, the broader UCP framework aims to standardize how AI handles checkouts across different platforms. By partnering with Flipkart to develop these agentic capabilities, Google is building the infrastructure necessary to turn Gemini into a comprehensive shopping assistant that can execute tasks on behalf of the user.

Market dynamics and the Walmart connection

The choice of Flipkart as a primary partner highlights the deep financial and strategic ties between the involved players. Flipkart is owned by Walmart, and Google itself holds a minority stake in the platform following a funding round in 2024 where the search giant invested approximately $350 million.

This partnership creates a powerful ecosystem, but it also highlights the competitive landscape of AI commerce. During the tests observed by TechCrunch, Amazon listings were still visible within Gemini and AI Mode, yet they lacked the direct Buy button. This suggests that the transactional layer of Gemini is currently a privileged feature reserved for specific partners, giving Flipkart a significant visibility and conversion advantage within the AI interface.

Google is always testing new features and experiences to help people discover and connect with businesses more easily.

The global race for transactional AI

Google is not acting in a vacuum. The push toward commerce integration is a central theme in the current AI arms race. Competitors like OpenAI are similarly striving to move their models beyond simple information retrieval. The industry is shifting toward a model where AI does not just answer the question What is the best smartphone for my budget? but instead handles the subsequent action: Buy that smartphone for me.

By testing these features in India—a market with high mobile penetration and a sophisticated digital payment infrastructure—Google is using the region as a laboratory for the future of global e-commerce. The success of the Flipkart integration will likely dictate how Google rolls out similar transactional features in the US and UK markets.

Global implications for businesses in USA and UK

For entrepreneurs and retailers in the US and UK, the India-Flipkart test is a blueprint for the near future of customer acquisition. When AI assistants move from discovery to transaction, the traditional sales funnel is compressed. The click-through rate (CTR) from a search result to a website may become less relevant than the conversion rate within the AI agent itself.

In the US and UK, where antitrust scrutiny regarding Google's search dominance is intense, the rollout of such features will likely be more cautious than in India. However, the introduction of the Universal Commerce Protocol suggests that Google may attempt to avoid monopoly accusations by offering an open standard that any retailer can adopt, rather than a closed system. Businesses should prepare for a shift where optimizing for AI agents (Agent Engine Optimization) becomes as critical as traditional SEO. The ability of a product to be transaction-ready for an AI agent will soon be a primary competitive advantage in the global digital marketplace.

FAQ

---

Sources: TechCrunch, Indianexpress, Semasocial ·