The Next Click Might Not Be Yours

For years, online shopping has followed a familiar pattern.

You search for a product, compare different options, decide what to buy and complete the purchase yourself. AI has often been there to recommend products or answer questions, but the final click has always belonged to you. That may be starting to change.

Visa and Animoca Brands have completed a live pilot in which AI agents searched for products and completed purchases on a customer’s behalf. Rather than simply recommending what to buy, the technology carried out the transaction itself, moving agentic commerce from concept into real-world testing.

That changes where trust sits.

We’ve become used to AI helping us make decisions, allowing it to act on those decisions is something different. Instead of choosing every product individually, people may begin setting preferences, budgets and permissions in advance, allowing AI to handle routine purchases while they retain overall control. But that also raises new questions;

• Which purchases should an AI agent be allowed to make?
• When should it ask for approval?
• If something goes wrong, who is responsible?

As agentic commerce develops, trust, permissions and consumer protection may become just as important as the technology itself.

The technology is still in its early stages, but it marks an important shift that AI is no longer being developed simply to recommend the next purchase, the next click might not belong to the customer at all.