The Next Customer on the Internet May Not Be Human

If AI agents become part of purchasing decisions, the rules of e-commerce, marketing, and payment systems could change.

One of the internet's fundamental assumptions that we have rarely questioned so far is that purchasing decisions are always made by humans. We would realize we had a need, search for it, compare options, and ultimately make a choice. Search engines were built on this behavior, e-commerce sites were designed around it, and digital advertising and loyalty programs also developed from the same assumption.

Technology changed over the years. The internet got faster, mobile apps entered our lives, and AI-powered recommendation systems became widespread. But amid all these changes, one thing remained constant: the decision-maker was human.

Today, when we want to buy a product, we are still the ones who research, evaluate, and make the final choice.

But what if this starts to change?

What happens if, a few years from now, the most active customers on the internet are not humans themselves, but AI agents acting on their behalf?

At first glance, this question might seem a bit bold. After all, shopping, planning a vacation, or choosing a financial product seem like quite personal choices. Most of us assume we want to make these decisions ourselves. However, recent developments in AI technologies show that, even if not the entire decision-making process, certain parts of it could increasingly be handed over from humans to software.

Perhaps the real change in the coming years will not be AI giving more accurate answers, but AI beginning to make decisions on our behalf within certain limits.

What Changes If AI Starts Making Decisions?

Over the past few years, AI has become one of the most talked-about topics in our lives. Systems such as ChatGPT, Copilot, Gemini, Claude, and similar ones generate content, answer our questions, perform analysis, and increase our productivity in many areas.

However, to understand the concept of agentic commerce, we need to clarify the difference between the AI systems we use today and the new generation of agents.

Most existing systems focus on producing information. We ask a question and get an answer. We request an analysis and see an evaluation. The one who initiates the process, manages it, and makes the final decision is still the human.

Agents, however, stand at a different point. Viewing them merely as more advanced chat systems would be incomplete. That's because agents can pursue a specific goal, evaluate alternatives to reach that goal, communicate with different systems, and take action within predefined rules.

For example, if you tell an AI system today, “Find me an affordable flight to London next month,” you get various options and you make the final decision. In the agent approach, however, the system doesn't just present options; it evaluates alternatives by taking into account your travel dates, budget, preferred airlines, and past choices, and can complete the booking within the authority limits you've set.

This difference may seem small at first glance. But what actually changes here is not the technology, it's the decision-making mechanism itself.

To put it in a simple framework: generative AI answers questions, AI agents act toward specific goals, and agentic commerce is when that action turns into researching, selecting, and, when necessary, carrying out commercial transactions.

What makes this possible is not just AI models becoming smarter. Over the past few years, tool-using AI systems, API ecosystems, digital identity infrastructures, open banking platforms, and digital wallets have all started maturing at the same time. AI is no longer just a system that produces information; it is turning into an actor capable of carrying out specific transactions on the internet. This is exactly what makes agentic commerce significant. What's changing is not just what AI knows, but what it can do.

What Assumption Is Digital Commerce Built On?

Much of the e-commerce world is built on capturing people's attention and influencing their decisions. Companies try to drive traffic to their websites, optimize product pages, invest in advertising, and improve the customer experience. The purpose of all this effort is to be visible at the moment of decision and to steer the user toward their own product.

Google's growth, the rise of social media platforms, and the success of marketplaces were also largely built on this same logic.

Because the one making the decision was the user.

This was perhaps the most important assumption digital economics has been built on so far.

With agentic commerce, this assumption may need to be reconsidered. If a new decision layer enters between the user and the seller, the concept of visibility may take on a different meaning. It may not be enough for brands to simply persuade the customer; they may also need to stand out in the criteria of the systems that evaluate on the customer's behalf.

Although this may look like a small change at first glance, it actually represents a transformation that could affect how the digital economy functions.

What Happens If People Stop Searching?

Today, when we want to buy a new laptop, smartphone, or TV, we go through a similar process. We search, visit different sites, read user reviews, watch videos, compare technical specifications, and often make a decision only after hours of research.

In fact, we spend a significant portion of our time choosing products rather than using them.

One of agentic commerce's core promises emerges exactly at this point: reducing the user's research burden and managing complex decision processes on their behalf.

The user defines their need. The system researches, compares, filters, and identifies the best options. In certain cases, it doesn't just provide recommendations, it can also complete the transaction.

What's notable here is not speed.

What really matters is the point at which decision-making responsibility begins shifting from the human to the system.

If this model becomes widespread, it's not just the shopping experience that changes. The role of search engines could shift, advertising models could transform, the concept of SEO could take on new meaning, and the relationship brands build with customers could be redefined.

Because if the user stops searching, the rules of being visible change too.

Does This Change Also Affect Payment Systems?

Agentic commerce is often evaluated from an e-commerce perspective. Yet the effects of this transformation may not be limited to product selection alone, because the payment process inevitably follows the purchasing decision.

An AI agent can research products.

But can it make the payment?

Within what limits will it operate?

Which payment instrument will it use?

Who will grant this authority?

If an incorrect or unauthorized transaction occurs, who will bear the responsibility?

The payment systems we use today were largely designed on the assumption that the party initiating and approving the transaction is human. Agentic commerce questions exactly this assumption.

For this reason, the topic isn't just about AI. It is also directly connected to authentication, authorization, open banking, digital wallets, card payment systems, security models, and financial liability mechanisms.

I won't try to answer these questions in this article. But it's fair to say that a significant portion of what banks, fintech companies, card schemes, and digital wallet providers will be discussing in the coming years will take shape around these topics.

Because what changes in the future may not only be the shopping experience, but the payment experience itself.

In the Future, Whose Trust Will Brands Try to Win?

In the early days of the internet, competition was shaped around access to information. In the mobile world, user experience took center stage. In the platform economy, customer relationships and customer ownership became among the most critical assets.

In the age of agentic commerce, the new battleground for competition may be different.

Today brands compete to capture customers' attention. Tomorrow, they may need to compete to stand out in the selection criteria of the systems that evaluate and choose on customers' behalf.

Perhaps the most valuable companies of the future won't be the ones that appear most on the customer's screen, but the ones most preferred by the systems acting on the customer's behalf.

This means not only commerce but also marketing, customer loyalty, and the digital platform economy being reshaped.

In today's world, competition is largely built on the attention economy. Tomorrow, attention may be replaced by trust, and trust, in turn, may be replaced by having an advantage within agents' evaluation mechanisms.

What Is the Real Question We Should Be Asking?

At the end of the Closed-Loop Payment Systems series, we arrived at this question:

Whose ecosystem does the customer stay in?

With agentic commerce, a new question is added to this one:

Who is making the decision in the customer's place?

Although this question may look technological at first glance, it is actually an economic one. Because when the decision-making mechanism changes, the economic model built around it has to change as well. From search engines to payment systems, from loyalty programs to digital marketing, a significant portion of today's internet economy is built on the assumption that humans make the decisions.

If the decision-making party increasingly shifts from humans to AI agents, the rules of the internet economy may be rewritten as well.

In the next installment of this series, I will look at how this transformation could change one of the internet economy's most fundamental behaviors: searching.

Because if users stop searching, the internet economy could change much faster than we think.