Google Is Testing AI Shopping on Flipkart: Gemini Is Getting Closer to the Buy Button
I’ve always thought the strangest part of online shopping was that the internet already knows how to find almost anything, yet the final journey still involves a pile of tabs, filters, product pages and checkout screens.
Google wants to remove more of that friction.
The company is testing a new experience in India that lets selected users see a Buy button on certain Flipkart products inside Gemini and Google AI Mode.
That sounds like a small interface change.
It is not.
The experiment is another step in Google's larger plan to move AI from shopping assistant to shopping participant—a shift that could eventually affect how Americans search for products, compare prices and complete purchases.
This Is Not Just a Flipkart Feature
The obvious headline is that Gemini can now help some Indian shoppers buy from Flipkart.
The more important story is what Google has been building underneath.
In January, Google introduced the Universal Commerce Protocol, or UCP, as an open standard for connecting AI agents, merchants and payment providers across the shopping journey.
Google designed UCP so an AI agent can move from discovery to product information, checkout and eventually post-purchase support without requiring a custom technical relationship for every individual AI system.
That is the infrastructure that makes agentic shopping practical at scale.
What Happens When Gemini Finds a Flipkart Product?
The current experiment keeps the experience relatively simple.
An eligible Flipkart product appears in Gemini or AI Mode. Test users can see a Buy button, and pressing it brings up a Flipkart-branded checkout experience within the AI interface.
Other users can still see Flipkart products in Gemini and AI Mode but do not necessarily receive the purchase option.
This distinction is important because Google has not announced that every Gemini shopping result can now be purchased directly.
Google Already Has a Bigger Shopping System Behind Gemini
Google says its Shopping Graph contains more than 60 billion product listings.
That database combines product information with signals such as inventory, pricing and reviews, giving AI systems a much richer foundation than a conventional web search index.
Google also says people shop across its products more than a billion times a day.
That scale explains why Google is investing heavily in AI commerce.
The company already owns the discovery surface, the shopping data layer, the payment infrastructure and the Gemini AI platform.
The missing piece has been connecting all four.
Universal Cart Is the Bigger Clue
At Google I/O 2026, Google introduced Universal Cart as another major piece of the puzzle.
Google describes it as a shopping cart that works across different Google surfaces, including Search and Gemini, with plans to connect other services as well.
Products can be added while shoppers browse, and Google says the system can help identify price changes, availability and compatibility issues.
When it is time to purchase, UCP can support checkout through Google Pay for participating merchants or transfer the customer to the retailer's own website.
The important architectural shift
Google is no longer treating shopping as one isolated Search feature.
It is building a layer where discovery, comparison, cart management, payment and retailer interaction can be coordinated by AI.
India Is More Than a Convenient Test Market
India is particularly useful for this experiment because Flipkart already operates at enormous scale.
Walmart says the Flipkart Group has more than 500 million registered users and offers more than 150 million products across more than 80 categories.
That gives Google a huge real-world environment in which to observe how people respond when conversational AI becomes part of the purchase path.
Google's own India research adds another interesting detail.
Google says consumers who use AI features navigate an average of 13 touchpoints before buying, compared with 5.5 for non-AI users in its cited research.
That suggests AI is not necessarily making shoppers rush through the decision.
It can actually create a more extensive research journey before the final purchase.
Why U.S. Shoppers Should Pay Attention
The Flipkart test is happening in India, but the underlying technology is not confined there.
Google's UCP ecosystem was developed with major retail players including Walmart, Target, Shopify, Etsy and Wayfair, and Google has described checkout experiences across major U.S. retailers.
That makes the India experiment useful as a preview of where Google's U.S. products could go.
For an American shopper, the eventual experience could look less like “search Google, click Amazon, compare products, open another tab” and more like a conversation.
You tell Gemini what you need.
The AI narrows the choices.
You approve the product.
The transaction happens through a connected merchant or payment flow.
Google wants that entire sequence to feel like one experience.
But the Buy Button Creates a New Problem
Recommendation mistakes are annoying.
Purchase mistakes cost money.
That difference becomes critical when AI moves from helping users decide to participating in transactions.
Reuters reported this week that major banks are warning about risks from AI shopping agents, including fraud, privacy breaches, incorrect purchases and the possibility of agents steering users toward less-protected payment methods.
The banks also called for clearer disclosure when AI is involved in a transaction and greater transparency around how shopping agents make decisions.
The rule for consumers should be simple
Let AI do the research. Keep explicit control over the final purchase, especially when the order is expensive, recurring or difficult to reverse.
Google's Payment Architecture Matters
Google's UCP work is being developed alongside the Agent Payments Protocol, or AP2.
The objective is not simply to let an AI type credit-card details into a website.
Google has described a system in which payment authorization can be represented securely and controlled by the user.
This distinction becomes essential as AI agents become more autonomous.
A shopping agent should be able to discover an interesting laptop without automatically having permission to spend $2,000 on it.
There Is Also a Subtle Business Question
AI changes how products get discovered.
In traditional search, retailers compete for a position on a results page.
In AI shopping, the system may recommend only a few products after interpreting the user's needs.
That concentrates enormous influence inside the AI's selection process.
Google has already introduced Merchant Center tools designed to help businesses understand how their products perform on AI surfaces.
That tells us something important: being visible to AI is becoming a distinct commercial problem.
The new retail optimization layer
Product titles, structured attributes, inventory, shipping information, reviews, promotions and merchant reputation can increasingly affect whether an AI system has enough reliable information to recommend a product.
That means the future of SEO may increasingly overlap with AI-readable commerce data.
Google and Flipkart Have More History Than This Test
There is another detail worth knowing.
In 2024, Alphabet's Google agreed to invest about $350 million in Flipkart for a minority stake, while also pursuing a Google Cloud collaboration with the company.
That investment does not establish why the current shopping experiment exists.
But it does show that Google and Flipkart have had a strategic relationship for years rather than suddenly becoming partners because of Gemini.
What Generic Coverage Often Misses
1. The real product is the commerce layer
The Flipkart Buy button is just the visible interface. UCP, product feeds, shopping data and payment infrastructure are what make the experiment possible.
2. AI could increase research before purchase
Google's India research suggests AI-using shoppers can travel through more touchpoints before buying, not simply fewer.
3. Merchant data is becoming AI infrastructure
Retailers need accurate product, inventory and shipping information because AI systems increasingly depend on structured commerce data.
4. Checkout is the trust boundary
Getting an AI recommendation is one thing. Giving an AI agent authority to spend money requires far stronger controls.
5. The India test provides a preview, not a guarantee
What Google learns from Flipkart can influence future products, but it does not mean the same interface will immediately appear for every U.S. retailer.
Google Gemini vs. Traditional Online Shopping
| Stage | Traditional Shopping | AI-Assisted Shopping |
|---|---|---|
| Discovery | Keyword search and retailer browsing | Natural-language request |
| Research | Open multiple pages and compare manually | AI summarizes and narrows choices |
| Cart | Retailer-specific | Can potentially span connected services |
| Checkout | Retailer website or app | AI-connected checkout or retailer handoff |
| Authorization | User initiates purchase directly | Requires explicit controls for agentic purchases |
Pros and Cons of AI-Powered Shopping
Potential Benefits
- More natural product discovery
- Faster comparison of complex specifications
- Less repetitive retailer browsing
- Potentially smoother checkout journeys
- AI can monitor prices and product availability
Important Trade-Offs
- AI recommendations can be inaccurate
- Merchant visibility may become more opaque
- Purchase errors have direct financial consequences
- Privacy risks grow as agents gain more shopping context
- Retailer and user support responsibilities remain evolving
Sundar Pichai Already Explained the Direction
“I’m looking forward to the day when agents can help me get from discovery to purchase.”
— Sundar Pichai, Google CEO, speaking about the future of agentic commerceThat sentence is almost a blueprint for the Flipkart experiment.
Google is trying to make discovery and purchase part of one continuous AI experience.
The technology is not finished, and the current test is narrow.
But the direction is remarkably clear.
Watch: How Google's Universal Commerce Protocol Works
This technical walkthrough explains Google's Universal Commerce Protocol and how it is designed to connect AI agents with merchants and checkout systems.
Amazon: Devices to Explore for AI-Powered Shopping
Google Pixel 11 Pro
Google's current Pixel flagship is a natural device for exploring Gemini-centered mobile experiences, including AI-assisted search, research and shopping workflows.
Check Pixel 11 Pro on Amazon →Samsung Galaxy S26 Ultra
Samsung's current flagship platform also puts AI-assisted features directly into the smartphone experience, making it useful for people exploring the growing overlap between mobile computing and AI assistants.
Check Galaxy S26 Ultra on Amazon →What Happens Next?
The immediate milestone is the broader Flipkart rollout reportedly planned for October.
After that, the more interesting question is whether Google can reproduce the experience across different merchants, categories and payment systems.
That is where UCP becomes important.
A protocol designed for one retailer is useful.
A protocol that lets an AI agent communicate with thousands of merchants is potentially transformative.
Google is clearly building toward the second scenario.
The Bigger Picture
The Flipkart test is easy to describe as “Gemini can now buy things.”
That description is too simple.
What Google is actually testing is whether an AI assistant can become the layer connecting a shopper's intent to a merchant's transaction system.
Today, that layer is limited to selected Flipkart products and users in India.
Tomorrow, it could become a normal part of Google Search, Gemini and other Google surfaces.
The real competitive battle will then move beyond who has the best chatbot.
It will be about who controls the interface where consumers discover products, make decisions and authorize purchases.
Google is already building that interface.
The Flipkart experiment is simply one of the first places where we can see the pieces coming together.
Physical AI 2026: Separating Hype from Reality
As AI moves beyond screens and into humanoid robots and autonomous factories, separating trillion-dollar market hype from deployable reality is critical. Read our complete 2026 Physical AI guide to explore NVIDIA's robotics stack, the progress of Tesla Optimus, and the actual state of the market.
Read the Physical AI Guide →Sources and further reading:
TechCrunch — Google tests buying from Walmart-owned Flipkart through Gemini and AI Mode in India
The Indian Express — Google tests Flipkart Buy option within Gemini and AI Mode
Google Developers Blog — Under the Hood: Universal Commerce Protocol
Google — Introducing Universal Cart and more ways to help you shop
Google — New technology and tools for retailers in the agentic-shopping era
Google — Win AI Search-Powered Shoppers in India
Walmart — Flipkart Group in India
Reuters — Banks warn AI shopping bots raise scam, fraud and data-privacy risks
Frequently Asked Questions About Google Gemini and Flipkart Shopping
Can Gemini buy products from Flipkart?
Google is testing that capability for selected users in India. Eligible Flipkart listings can display a Buy button inside Gemini and Google AI Mode, leading into a Flipkart-branded checkout experience.
Is Gemini shopping through Flipkart available to everyone?
No. The current experience is a limited test covering selected users and a small number of product categories. Broader availability has been reported for October, but Google has not announced universal access.
What is Google's Universal Commerce Protocol?
Universal Commerce Protocol, or UCP, is an open standard designed to let AI agents, merchants and payment providers communicate across the shopping journey, including product discovery, checkout and other commerce functions.
Does UCP mean Gemini can buy anything automatically?
No. UCP is an infrastructure standard, while individual shopping experiences still depend on merchant participation, product eligibility, payment systems and user authorization.
Will AI shopping eventually reach U.S. retailers?
Google is already building UCP-powered commerce experiences with major retailers and has described checkout through Google surfaces including Gemini and AI Mode. The exact merchant coverage and availability in the United States will continue to evolve.
No comments:
Post a Comment