For decades, technology served the shopper. Now, increasingly, it shops for them. The question is whether we are ready for what that means.
Think about the last time you genuinely decided what to buy. Not reached for something on a shelf, or typed a search query into a box, but actually sat down and made a considered choice about a product, a price, a brand. For many consumers today, that kind of deliberate decision is becoming rarer. Recommendations arrive before the question is asked. Carts fill themselves. Subscriptions renew without prompting. Orders are placed and sometimes dispatched before a person has consciously registered they needed something at all.
Founder & CEO
Nitro Commerce
This is not science fiction, and it is not a distant prospect. It is the direction commerce is already moving, and the pace is accelerating faster than most people have noticed. Artificial intelligence, for a long time framed as a helpful tool in the background of online shopping, is taking on a fundamentally different role. It is moving from advisor to decision-maker. And the implications of that shift, for consumers, for businesses, and for regulators, deserve far more attention than they are currently getting.
From Suggestion to Action
There is a useful way to think about the stages of AI involvement in commerce. The first stage was recommendation: the algorithm looks at what you have bought and suggests what else you might like. This is where most people assumed the story ended. But it did not. The second stage was personalization at scale: pricing, product display, and marketing all shaped in real time around what the system knows about you individually. The third stage, which is now well underway, is something different. This is agentic AI, systems that do not just advise but act.
An agentic AI system in commerce does not wait for a consumer to browse. It monitors behaviour, anticipates need, evaluates options across sellers and prices, and completes transactions on the user’s behalf. It works round the clock to manage the system end to end. The consumer sets preferences once and the machine handles the rest. In its most advanced form, the consumer is not really a shopper anymore. They are the principal, and AI is the agent carrying out their commerce on their behalf.
India is Not on the Sidelines of This Shift
It would be tempting to frame this as a primarily Western phenomenon, something driven by large American technology companies and distant from the day-to-day reality of Indian commerce. That framing would be wrong. India is, in several meaningful ways, at the centre of this transformation, and both its government and its private sector are moving quickly.
The IndiaAI Mission, approved by the Union Cabinet in March 2024 with a budget outlay of Rs 10,371.92 crore focuses on building the infrastructure that makes AI-led commerce not just possible but scalable. The mission has already crossed 38,000 GPUs in its computing infrastructure, well beyond its initial target of 10,000, providing affordable AI processing capacity to startups and enterprises across the country. This has enabled AI applications to be developed across all sectors majorly in e-commerce and retail.
Alongside this, India’s Open Network for Digital Commerce (ONDC) has grown at a pace that surprised even its early supporters. By November 2024, ONDC was processing 14.45 million transactions in a single month, up from 13.9 million in October. Non-mobility categories on the platform, covering food, grocery, fashion, and home goods, saw growth of 825 percent over eight months. As of August 2025, the network had expanded to over 616 cities, with more than 7.64 lakh merchants onboarded. The open, interoperable structure of ONDC, designed to work across buyer and seller applications, is precisely the kind of digital infrastructure that autonomous commerce systems need to function at scale.
What Autonomous Commerce Actually Looks Like in Practice
The practical applications are already visible, even if the vocabulary used to describe them has not caught up. Grocery platforms that auto-schedule weekly orders based on past buying patterns. Medicine delivery services that track prescription cycles and prompt refills before stock runs out. Food delivery apps where the suggested order is so accurate to the user’s habits that accepting it with one tap is easier than actually choosing. These are not hypothetical features being tested in labs. They are live, in use, and normalising a new kind of consumer relationship with the act of buying.
In the business-to-business context, the shift is even more pronounced. Retail supply chains are being managed by AI systems that predict demand at the level of individual store locations, generate purchase orders automatically, optimize vendor selection on price and delivery reliability, and adjust inventory positions in real time. Human procurement managers in these environments are increasingly reviewing decisions the system has already made, rather than making decisions themselves. The work has shifted from execution to oversight, and the boundary between the two is moving constantly.
India’s own e-commerce ecosystem reflects this clearly. According to NASSCOM AI Adoption Index, nearly 89 percent of startups launched in 2024 integrated AI into their products or services. In this index, India scores 2.45 out of 4, reflecting rapid enterprise-level integration. The country also ranks among the top four globally in AI skills and capabilities according to the Stanford Institute for Human-Centred Artificial Intelligence. These numbers are not just about talent. They reflect an economy that is actively building the systems that will define how commerce works in the next decade.
The Questions That Come with This Territory
None of this is without complication. When AI makes buying decisions on behalf of a consumer, questions of accountability become genuinely difficult. If a system purchases a product that causes harm, who is responsible? The consumer who set the preference? The platform that built the agent? The seller whose product was selected? These are not abstract legal puzzles. They are practical questions that existing consumer protection frameworks were not designed to answer.
There is also the question of what happens to consumer choice when the system doing the choosing has a commercial interest in particular outcomes. Every major platform that has moved toward AI-assisted or autonomous commerce has its own business model, its own preferred sellers, its own margin calculations. The algorithm that decides what goes in your cart is not a neutral referee. It is an actor with incentives, and those incentives do not always align with the consumer’s best interest.
India’s government has moved faster on this than many anticipated. In November 2025, the Ministry of Electronics and Information Technology released its AI Governance Guidelines under the IndiaAI Mission, a framework that takes a deliberate, risk-based approach to AI deployment across sectors. Rather than rushing an umbrella AI law, the guidelines assign oversight to sectoral regulators, the Reserve Bank of India for fintech applications, SEBI for securities, and so on, while establishing core principles around transparency, accountability, and the protection of individuals. The framework explicitly addresses the question of AI systems that generate decisions autonomously, and whether existing legal definitions need to be updated to account for them.
The Bigger Picture
It is worth stepping back from the technology for a moment and asking what kind of relationship between people and commerce we actually want. The efficiency gains of autonomous commerce are real. Time saved, costs reduced, friction removed. For a consumer managing a busy life, having a system handle routine purchases intelligently is genuinely useful. Nobody mourns the hours spent writing shopping lists.
But commerce has always been more than a transaction. It is how people express preferences, compare options, support local businesses, and exercise judgment. When those functions are progressively delegated to machines, something changes in the nature of economic agency. It is not necessarily a change for the worse. But it is a change that deserves to be made consciously, with clear rules about what AI can decide, what it must ask, and what remains firmly in human hands.
India stands at a genuinely interesting point in this story. It has the infrastructure ambition, the talent base, the digital commerce network, and now an emerging governance framework to shape how autonomous commerce develops on its own terms. The decisions made over the next few years, about what AI is allowed to do in commerce and what it is not, will define the experience of hundreds of millions of consumers for a generation. That is not a technology conversation. It is a values conversation wearing a technology hat. And it is one that needs to happen now, before the defaults are set so deep that changing them becomes practically impossible.
Authored by Umair Mohammed, Founder & CEO at Nitro Commerce