AI Agents Transform Ecommerce as Agentic Commerce Reshapes How We Shop Online

Agentic commerce: why AI agents are transforming the ecommerce landscape

AI agents are transforming e-commerce through autonomous transactions Digital commerce has historically focused on optimizing the customer experience through interface design, structured navigation, and carefully engineered conversion processes. Over time, merchants have improved these experiences to reduce friction and support decision-making at every step of the purchasing lifecycle. While AI historically played a limited role, restricted to basic chatbots or shopping assistants, rapid advancements mean AI agents are now prepared to execute transactions directly on behalf of customers. This shift is driving a fundamental transformation across the ecommerce sector. This new ecosystem is agentic commerce, where software agents initiate and complete transactions within clearly defined constraints. To accommodate this, payment infrastructure is moving closer to the point where intent becomes execution, introducing protocols that make platforms machine-readable so AI can safely complete transactions in real time. It is a major leap forward for retail technology, and enterprise readiness is urgent. The Enterprise Blueprint: Brand vs. Non-Brand Agents For large enterprise merchants, this shift introduces both an immediate opportunity and a critical defensive necessity. Enterprise adoption will likely begin on a merchant's own digital properties through brand agents, which are dedicated AI assistants designed to improve conversion, capture valuable data, and keep the consumer firmly within the merchant's ecosystem. Eventually, these properties must open up to external, non-brand agents controlled by consumers or procurement departments. As AI agents become the primary interface for ecommerce, enterprise merchants who fail to make their platforms discoverable and transactable risk losing market share to competitors who are ready. Securing the Payment Layer, Navigating Discovery To capture these autonomous sales, a merchant's infrastructure must interact seamlessly with software systems. When autonomous agents handle procurement, they process data directly rather than navigating traditional user interfaces. While optimizing product catalogs and metadata for LLMs is vital for discovery, enterprise merchants do not need to tackle this layer alone; they can solve this through specialized discovery and platform partners. The core operational challenge for the merchant remains the payment layer. Because AI agent activity makes transaction volumes highly dynamic, minor inefficiencies or a single failed authentication step can terminate an entire chain of transactions. The priority for merchants is establishing a robust payment architecture capable of verifying human intent and explicit consent, recognizing and authenticating the specific AI agent, and processing transactions securely across multiple rails in real time. Shifting to Modular Business Models The applications of agentic commerce vary across sectors due to distinct transaction models, regulatory systems, and the structural maturity of different digital verticals. However, a recurring theme is the transition from rigid, packaged bundles to highly granular, modular transactions. For example, instead of requiring a fixed monthly or annual subscription for software-as-a-service (SaaS), an AI agent can dynamically subscribe a user to a platform precisely when needed. The agent continuously assesses usage and pays for the appropriate tier automatically. This allows subscription models to match real-time demand, aligning perfectly with customer utility. Similarly, online learning platforms can deploy micro-transaction frameworks. Rather than purchasing full courses, users can access a single lesson or group of lessons at a bespoke price point. Agents can combine individual lessons from multiple providers to create a tailored learning experience, while the underlying payment infrastructure fragments and distributes the value seamlessly across all accessed merchants. Programmable Monetary Flows At the heart of agentic commerce is the transition from traditional payment infrastructure to programmable monetary flows. In this environment, systems execute transactions based on continually evaluated conditions of intent and permission. Confirming human consent is vital, as it serves as the primary defense protecting merchants from claims of unauthorized or fraudulent AI activity. To enable this, payment environments must interpret delegated instructions, enforce spending constraints, and execute transactions instantly. Agent-bound payment credentials facilitate these purchases, allowing agents to act autonomously while preserving financial control and traceability for the end user. Global card schemes are already formalizing the components of this model, signaling a broader evolution of delegated payment frameworks. Security infrastructure must also evolve to counter malicious actors attempting to counterfeit legitimate agent behavior. Regulatory updates, such as the upcoming PSD3/PSD4 frameworks in Europe and the EU AI Act, are setting clearer expectations for accountability, making compliance more critical than ever. From Complexity to Simplification Navigating this new terrain involves balancing real-time agent authentication, compliance with shifting regulations, and programmable credentials, all of which introduce undeniable operational complexity. But preparing for it doesn't mean overhauling your existing systems. The path forward lies in a single integration point that abstracts this protocol churn away from your business. By partnering with the right payment layer expert, enterprise merchants can simplify the complex backend architecture, shielding their operations from technical friction while ensuring they are ready to accept machine-to-machine payments seamlessly. The protocol-level reset for digital trade is already accelerating. Competitive advantage belongs to organizations that secure their payment infrastructure early, start by assessing whether your payment architecture can verify intent and authenticate agents today. We list the best mobile credit card processors. This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

The agentic commerce gold rush risks repeating ecommerce's biggest mistakes

Agentic commerce has become the latest obsession in ecommerce. The technology is certainly compelling. Instead of consumers browsing websites, comparing products and making purchases themselves, AI agents increasingly do the work for them. They search, compare, recommend and transact on our behalf. According to the hype, agentic commerce has the potential to transform the way we shop online. It is easy to see why merchants are excited. For an industry under constant pressure to drive growth, improve efficiency, capture shoppers' attention, and deliver better customer experiences, agentic commerce promises a great deal. Some believe it could become as significant as the shift to mobile commerce. Others see it as the next logical evolution of ecommerce itself. Whether those predictions prove accurate remains to be seen. New technologies often arrive surrounded by hype. Not all of them live up to expectations. But retailers are taking agentic commerce seriously. Many are moving rapidly to integrate it into their businesses. No one wants to be left behind when a new technology goldrush gathers momentum. That urgency is understandable. But it should also give us pause for thought. Every new development in ecommerce has brought with it new opportunities for fraud. Online payments made shopping more convenient but opened the door to card-not-present fraud. Mobile commerce transformed how consumers shop but introduced new security challenges. More recently, AI has begun reshaping how businesses work while simultaneously giving fraudsters powerful new tools to scale attacks. There is little reason to believe agentic will be any different The pattern is familiar: innovation creates opportunity, fraudsters adapt, and businesses scramble to catch up. Agentic commerce has all the ingredients to follow a similar path. Part of the reason adoption is moving so quickly is that ecommerce growth has become harder. Competition is intense and customer acquisition costs remain stubbornly high. No wonder many merchants are searching for new ways to improve efficiency. Early adoption of AI agents appears to offer the golden ticket solution. They promise faster transactions, less friction and greater automation at a time when growth is harder to achieve. The challenge is ensuring that risk management keeps pace. Are we trusting AI agents too much? One of the more interesting aspects of the agentic commerce debate is the level of trust many businesses are already placing in AI systems. That may sound surprising until you consider the reality facing many merchants today. Customer fraud continues to grow. Refund abuse, chargeback fraud, promotion abuse and policy exploitation have become major challenges for ecommerce businesses. Our own extensive polling of merchants shows many place greater trust in the promise of AI agents than in their own customers. On the face of it, AI agents are attractive. They appear predictable. They follow instructions. They don't deliberately manipulate systems for personal gain. But it's crucial to understand replacing one source of risk with another is not the same as eliminating risk altogether. AI agents may not behave like human fraudsters, but they introduce new vulnerabilities that businesses are only now beginning to understand. More on that shortly. The accountability problem One of the biggest unanswered questions surrounding agentic commerce centers on liability. If an AI agent makes a poor purchasing decision, where does accountability sit? If a fraudulent transaction takes place, who is liable? If an agent is manipulated into making purchases it shouldn't make, who is responsible for the resulting losses? The consumer? The retailer? The technology provider? The operator of the agent itself? The problem is existing frameworks were built around human decision-making. Agentic commerce introduces a new layer of autonomy that blurs traditional lines of responsibility. At the moment, many of these questions remain unresolved. That doesn't mean businesses should avoid agentic commerce. But it does mean they should think carefully about governance, oversight and accountability before deploying it at scale. Trust cannot simply be assumed because a transaction is being conducted by an AI agent. Merchants need visibility over who is operating that agent, how decisions are being made and what safeguards exist when things go wrong. Fraudsters will move quickly If history teaches us anything, it is that criminals are among the fastest adopters of new technology. While businesses focus on the opportunities created by innovation, fraudsters focus on the weaknesses. Agentic commerce is likely to create plenty of both. We will see attempts to hijack legitimate agents and manipulate their behavior. Fraudsters could also create fake agents that impersonate trusted services or brands. Criminal networks may also deploy their own autonomous agents to identify vulnerabilities and exploit them at scale. Promotions, loyalty schemes and refund processes will almost certainly also become targets for increasingly sophisticated forms of automated abuse. What makes this new environment particularly challenging is the speed of evolution. Most forms of online fraud still require a degree of human input. Autonomous agents have the potential to operate around the clock, making decisions, testing vulnerabilities and exploiting weaknesses at machine speed, as well as adapting their tactics automatically when they fail. The result may not be more fraud in absolute terms, but we will see more sophisticated and fast-moving threats. That will require merchants to rethink how they identify risk, monitor activity and respond. Proceed, but with your eyes open None of this should be interpreted as an argument against agentic commerce. The technology has genuine potential. It could make shopping easier, reduce friction and unlock new opportunities for both consumers and businesses. But innovation and risk have always evolved together. The companies that succeed in the next phase of ecommerce will not necessarily be those that move first. They will be the organizations that understand the risks as well as the opportunities. That means understanding their own data. Maintaining visibility over customer and transaction behavior and continuously monitoring for unusual activity means controls can be adapted as new threats emerge. In a world where both shoppers and fraudsters may increasingly be represented by AI, capturing diverse data at scale and knowing how to interpret it to protect against online crime and serve good customers better will become more important than ever. We've featured the best IT automation software. This article was produced as part of TechRadar Pro Perspectives, our channel to feature the best and brightest minds in the technology industry today. The views expressed here are those of the author and are not necessarily those of TechRadarPro or Future plc. If you are interested in contributing find out more here: https://www.techradar.com/pro/perspectives-how-to-submit

Know Your Agent - Experian's position on trust for AI commerce

Cast your mind back to the early to mid-90s, and as the Internet burst into public consciousness, it seemed hard to imagine that this virtual platform would be where you would shop for groceries, clothing, travel, insurance or services from the local government. The thought of carrying out banking online, or through a similarly newish mobile phone seemed preposterous. Yet today that is exactly what each and every one of us does. Pure online banks deliver a far better service than the old stalwarts of high and main street. A similar moment of system change is upon us once again; transacting through AI seems odd. Why? Because the same obstacles that faced the Internet back in the 90s currently exist for AI: there is a lack of trust, but a group of financial and technology services companies is betting that just as the Internet transformed shopping, so too will AI. Heeding the lessons of the Internet years, they have come together to build trust in AI-based transactions. Chief Innovation Officer of Experian, Kim Slaughter explains the new offering and why Experian is betting that history will repeat itself. AI agents carrying out transactions is perhaps the next logical step for the development of agentic AI. Business advisory firm Bain & Company forecasts that it will be worth between $300 billion and $500 billion in the US alone by 2030. That seems inflated, but then what isn't in the AI-era. In an interview following the release of Agent Trust, Slaughter at Experian says: We know there will be a shift. It will take some time to get there, and we haven't started to see the speed bumps ahead yet. Experian and its partners in Agent Trust, payments firm Visa, content delivery network (CDN) specialist Cloudflare and AI security provider Skyfire, though, are looking to get ahead of the curve. Slaughter believes the accelerated nature of AI will see the rise of agentic commerce arrive faster than many predict. She says her personal opinion is: 2027 will be when it takes off; there will be some interesting spikes as a result of the comfort we already have with AI. So the next step is whether people will want the additional convenience. She doesn't believe complex transactions like travel will be handed to AI agents, but a hot consumer trend could tip the balance to the agents: There will be something that everyone wants, and a lot of consumers will want an agent to find it, buy it, and don't disturb me by asking. Then there will be instructional videos on TikTok. Slaughter says this is the reason its new Agent Trust service is needed: As shoppers become comfortable with AI, they are not so comfortable with handing over banking credentials to AI and allowing agents to purchase for us. There have been a number of happy path demonstrations of agentic commerce where one transaction has gone through; there really hasn't been mass adoption. So we started talking to the end-to-end players, the merchants, card networks, banks, and consumer groups about what is preventing this from taking off, and we heard over and over again - trust. The merchants don't trust the agent traffic that is showing up at their door, because historically some of these bots were bad, and now they are not so bad. Consumers are not sure they can trust giving their credit card information over unless they are in the loop. She adds that the weak points of e-commerce are dogging agentic commerce, citing shipping of the wrong product, or arriving too late. Despite these problems, she says the customer is now confident that there is a way to remediate these issues through established processes. She adds: All this relies on knowing the humans and what was their intent and we realized that verifying humans is something that Experian has been doing for decades. Experian describes Agent Trust as a human-to-agent binding service, creating a Know Your Customer (KYC) service for agents, dubbed Know Your Agent (KYA). She adds: I believe there are a lot of similarities to mobile commerce; initially there was a concern of the card not being present. Now we are talking about the human not being present, so the whole ecosystem needs to adjust accordingly. Fraudsters will be quick to adopt agentic commerce as a way to rip businesses off, which Slaughter says is another motivation for the partners to create the know your agent service. AI is already the favored tool of cyber criminals who create deepfake identities. Meanwhile, the LLM providers, she says, do not believe the responsibility of remediating a scam is theirs. Agent partnership Experian and its partners have come together to offer five levels of assurance to the agentic commerce opportunity: the human-to-agent verification, a protocol that allows merchants to identify and verify an AI agent through a network from Visa, and Cloudflare protection. Together, these create an audit trail too. Slaughter adds: We verify the human when the agent is created. Their intent creates a cryptographic token that the agent carries in its header (the supplemental data sent before a payload), and we have tested this with Visa. She adds that having a CDN as part of the offer provides protection to the merchants, who can identify that an agent is verified and has good intent. The members of Agent Trust have delivered an open standards system that any LLM provider can adopt. Slaughter says initial agentic commerce initiatives from payments firm Stripe and ecommerce platform Shopify are closed ecosystems. The ability to analyze and manage bot traffic is becoming increasingly important to firms. AI bots provide a great hiding place for cyber criminals, and as we have reported, bot workloads on the estate are impacting the bottom line of firms, whose infrastructure takes a hit from bot crawling and may lose business to the AI sector. Slaughter adds: Merchants have beautiful store fronts and engaging content to make a buyer stay and buy more; well, the agent doesn't care, especially if it is just reporting back prices, and I think it is really going to become an issue for the larger players. My take Slaughter and I finished our conversation reminiscing on the smart homes that were demonstrated by the likes of Cisco in 2000. The connected fridge that orders staples remains as illusive as flying cars, but agentic commerce does have some potential when you factor in Internet of Things sensors. Maybe family staples will not trigger the agentic commerce revolution, but as we have seen with the Internet, we consumers do adopt technological commerce when the use case and the convenience stack up. Agent Trust is an example of lessons learned from the Internet era: trust needs to be won from the beginning. This will be a space to watch.

Is AI Buying Your Groceries?

Imagine not having to shop anymore because your "agent" does it for you. That's the pitch, at least. One of the 2026 buzzwords is agentic commerce, and if you've noticed a "Buy" button in a chatbot or heard your kids say they let AI handle their grocery reorders to save time, welcome to a new trend. Agentic commerce is real, growing quickly, and worth understanding before it appears in your digital shopping cart (or online account) with little fanfare or warning. The good news? Agentic commerce isn't as strange or risky as it sounds once you understand what's happening under the hood. Read:Best credit cards to obtain before retiring What is agentic commerce? For those of us like me (who loathe shopping and resent having to do it), this technology is akin to having your own competent, personal shopper who doesn't tire or get distracted - and who can compare 50 options in the time it took you to read this sentence. That's how the average AI shopping agent works. You tell it what you want - "find me a well-rated washing machine under $500" or "restock my typical weekly grocery order" - and it does the legwork. It will search, compare prices, check reviews, and in some cases, complete the purchase. The key word? Agent. This AI acts on your behalf, not unlike how travel agents book your flights and hotel stays without you having to make the calls yourself. You set the goal and the rules. The AI does the running around. If you think it sounds like some 25th-century sci-fi concept, it's not. It's already live. ChatGPT lets U.S. users buy directly from Etsy sellers without leaving the chat window. Also rolling out this year? The ability to shop at nearly unlimited Spotify stores, including brands like SKIMS and Vuori. Mastercard built Agent Pay, designed so AI agents to make verified purchases on your behalf. Visa has AI-ready cards built for agents. Perplexity has a Buy with Pro button connected to thousands of merchants through PayPal. The biggest names in payments and shopping have embraced this technology and trend, and are quickly building the infrastructure to support it. McKinsey researchers estimate AI agents could orchestrate up to $1 trillion in retail revenue by 2030. Where will you see agentic commerce? The short answer: many places. But these agents don't always announce themselves as AI shopping. Inside a chatbot You can ask ChatGPT or a similar tool for a product recommendation. In addition to links, a Buy button pops up in the conversation. You can confirm your shipping and payment information, and the purchase goes through without you having to visit the brand's website. As a set-it-and-forget-it reorder Wouldn't it be lovely to never run out of paper towels or cat litter again? An agent can reorder it for you when it notices you're running low (or because you've set a specific cadence, like Amazon and Chewy's subscribe options). You don't have to do a thing, unless you want to be asked first. Through a price-watching assistant You can tell an agent to buy something if the price drops below a certain dollar amount. It watches, waits, and completes the transaction as soon as the deal appears - which beats having to remember to refresh a webpage every day for a week. On retailer websites you already use Some online stores are redesigning their product pages to be easier for AI agents (not just humans) to read. You likely won't notice this part directly, but it's reshaping what appears when an agent goes searching on your behalf. Is Agentic commerce safe? This question matters the most, and it's the one companies are working hardest to answer. How do you set spending guardrails for an AI shopping agent? Three habits do most of the work here. Set guardrails immediately Before an agent buys anything, you tell it the rules: * A maximum price * Specific brands to stick to or avoid * Whether it should ask before buying or go ahead Industry groups are very clear on this point. Humans stay in charge of the goals and limits. The AI executes within them. You don't hand over your card number You don't give an AI agent your credit card number. The system creates a token, which is a stand-in number that only works for that specific purchase, with that specific merchant, for that specific amount. If something goes wonky, your actual credit card numbers aren't exposed. You confirm to finalize a purchase OpenAI's checkout system still requires you to tap to confirm the order, shipping address, and payment method before a charge goes through. You don't hand your wallet to a robot and walk away (at least, not yet - and only if you choose a more hands-off approach). The pros and cons of agentic commerce The upside: * It saves time. Comparing five retailers for the best price on a new mattress used to mean multiple open tabs and lots of back-and-forth. Now, you can share your specific requirements with a generative AI platform like ChatGPT, and it will return options in seconds. * It remembers things so you don't have to. Agentic AI remembers your sizes, preferences and what you bought last time because a good agent keeps track, like shop clerks once did. If you run a busy household, your AI assistant can place those automatic monthly orders for you. * It catches deals you might miss. Agents that watch prices for you, like airline and other big-ticket items, have their eyes open 24/7 (and are much more patient than humans refreshing a browser tab). The downside: * There's no consensus about what happens when something goes wrong. If an agent buys the wrong size or gets tricked into making a bad purchase, who's responsible? You, the company that built the agent, or the store? Researchers studying this phenomenon said there's no clear, universal answer yet. * Trust is still catching up to the tech. A 2025 Visa survey found that nearly 8 in 10 people interested in AI shopping expressed concern about data privacy. * It opens a new door for scammers. Wherever money moves automatically, fraud follows. Companies are building safeguards to reduce the likelihood that a cybercriminal will trick an agent with fake offers or misleading signals. But as with all tech, keeping ahead of scammers requires constant diligence. * It can nudge you toward purchases you didn't choose. If retailers pay agents to recommend their products, would you know? Currently, the process works on the honor system, and not all companies are committed to telling their customers. Should you try agentic commerce? There's no rush. But if you take advantage of Amazon's Subscribe & Save or Chewy's monthly autoship, you're already using a version of it. If your current way of online shopping works for you, agentic commerce won't replace it (and there are still plenty of folks - like me - who enjoy doing our own comparison shopping). If you're curious, start small. You're in control. Try it for something low-stakes, like restocking what you already buy each month. Keep the "ask me before you buy anything" setting on until you get a feel for how this technology works. In other words, treat it as you would a new assistant (or a toddler): give it a small task first. See how it does, and only hand over more control once you trust it to handle the smaller stuff. Before you try agentic commerce, check for these four things: The bots haven't taken over your grocery list entirely - yet. But they're getting really good at finding the best price on toilet paper, and for some people, that's a great thing to hand off. This story written for TheStreet by Nifty 50+ The Arena Media Brands, LLC THESTREET is a registered trademark of TheStreet, Inc. This story was originally published July 21, 2026 at 8:33 AM.