Agentic AI Is the Next Shift. Panic Is Optional.

Let’s go back to when we played music on our iPods and answered email on our BlackBerrys. Something bigger was quietly building in the background: the shift from brick-and-mortar shopping to online commerce. 

When commerce first moved online, it wasn’t just because we had to. It was because we wanted to. 

Yes, the shift from physical storefronts to the internet came with risk. Card-not-present transactions introduced new credit card fraud exposure. Shoppers became faceless. Trust had to be redefined through better fraud detection and transaction monitoring. But despite the uncertainty, businesses still leaned in. 

Why? 

Because online commerce represented growth. It represented reach. It represented opportunity. 

Going online wasn’t simply about enabling transactions through a new channel. It unlocked an entirely new way for people to engage with brands. Websites became destinations for discovery and research — places to compare options, explore products, read reviews, and build confidence before buying. Commerce expanded beyond the point of sale into a continuous relationship between brand and customer. 

Today, it’s almost impossible to imagine a successful company with no meaningful online presence. Not because every transaction must happen digitally, but because discovery, research, and engagement start there. Imagine where your business would be if you’d chosen to stay offline — if customers couldn’t find you, couldn’t research your products, couldn’t engage with your brand on their terms. 

The reality is simple: if a business isn’t discoverable, it effectively doesn’t exist for a large and growing segment of consumers. 

That same dynamic is beginning to shape agentic commerce. 

What Is Agentic Commerce? 

Agentic commerce is the emerging shift in which AI agents research, compare, and — increasingly — act on a shopper’s behalf, from narrowing options to initiating a purchase. 

It’s worth drawing a distinction here: most of what’s happening today is AI-influenced commerce, not fully autonomous commerce. A shopper might use an AI assistant to identify the right product, then go complete the purchase directly on the merchant’s site. Over time, more of that journey may be handed off to the agent itself — and as more decisions move from the agent’s recommendation to the agent’s action, the risk profile shifts right along with it. 

“By 2030, 20% of transactions will be executed through AI platforms — either via AI checkout or by AI agents.” 
Source: Gartner®, What Is Agentic Commerce?, March 2026, Sandy Shen. GARTNER is a trademark of Gartner, Inc. and/or its affiliates. 

The Parallel We Shouldn’t Ignore 

Agentic commerce feels disruptive because it introduces a new interface between shoppers and brands. Instead of a person clicking through a site, an intelligent agent may search, compare, and eventually buy on their behalf. That’s unfamiliar. And unfamiliar creates anxiety. 

But just like the early days of online commerce, this new channel is emerging because it offers real value. 

Shoppers are already using intelligent tools to research products, compare pricing, and narrow choices — part of a broader shift toward payment orchestration and automation that’s already reshaping how transactions get initiated. These behaviors are becoming part of how people decide, even while most transactions still close out on traditional ecommerce sites today. 

Agentic commerce isn’t just automation. It’s meeting people where they already are and supporting how they want to decide. 

57.6% of organizations report that 6% to 15% of transactions are already initiated by AI agents, underscoring the need for more sophisticated fraud controls as this channel scales. 
Source: Accertify Data 

The question isn’t whether this shift will happen. It’s whether brands participate thoughtfully — building fraud prevention and risk management into the channel from day one — or wait until the rules are written without them. 

Opportunity Comes With Responsibility 

No channel grows without risk. Online commerce taught the industry that trust doesn’t scale on its own — it has to be engineered. Fraud didn’t disappear. Chargebacks didn’t disappear. Instead, the ecosystem evolved: better fraud prevention software, sharper ai fraud detection, and stronger account takeover protection became table stakes for any business processing payments online. 

Networks, platforms, payment providers, and risk specialists didn’t solve those problems in isolation. They built shared standards, signals, and protections that let commerce grow safely. Agentic commerce will be no different — and the stakes are arguably higher. An AI agent acting on a shopper’s behalf can move faster than a human, which means transaction monitoring and identity verification need to be even more precise, not less. 

Accertify data reinforces the urgency: 92.8% of organizations expect agentic commerce to require strategy changes to their fraud prevention approach. That does not mean the ecosystem should slow down. It means trust, identity, and fraud prevention need to evolve alongside the experience. 

Source: Accertify Data 

That’s precisely where an adaptive, signal-rich approach to enterprise fraud management earns its keep. Static rules built for a human clicking “buy” won’t hold up against an agent transacting at machine speed — decisioning needs to adapt in real time, joining identity, behavioral, and transaction signals to tell a legitimate agent-driven purchase apart from a fraudulent one. 

This moment isn’t about moving fast and breaking things. It’s about moving forward deliberately, together. 

What Should Merchants Do to Prepare? 

Preparation doesn’t mean moving the fastest — it means engaging deliberately before agent-mediated activity reaches meaningful scale. A few things matter most right now: 

• Define what “agentic activity” actually means for your business — an AI referral, an agent-assisted session, and a fully agent-completed purchase are three different things, and treating them the same makes it hard to measure or manage any of them. 

• Start measuring what you can already see — referral sources, session behavior, and checkout patterns tied to AI tools — even before you can classify every interaction perfectly. 

• Treat agentic traffic as its own channel, not generic bot traffic to block and not conventional browser traffic to leave unexamined. 

• Strengthen identity and authorization infrastructure so you can establish who an agent represents and whether it stayed within what the shopper actually authorized. 

• Collaborate across the ecosystem — no single merchant will solve agent identification, consent, and liability questions alone, and the businesses that help shape those standards early will be better positioned than those that inherit them later. 

We Wanted This Once. And We Chose It Wisely. 

Agentic commerce is another chapter in a story commerce has told before. There’s no reason to panic — but every reason to prepare, with the same rigor that turned early ecommerce fraud prevention into a mature discipline over the last two decades. 

History shows the companies that engage early, help shape standards, and partner across the ecosystem are the ones best positioned to turn change into advantage. 

We see this moment the way we’ve always approached change in commerce: not as something to face alone, but as something the ecosystem solves together. Our role is to help clients build confidence through intelligence, experience, and collaboration — so they can say yes to more good commerce, even as the landscape evolves. That’s the thinking behind a fraud prevention platform built on adaptive intelligence: signals that get sharper with every transaction, whether that transaction starts with a human or an agent. 

We chose to go online because it represented opportunity. 

Agentic commerce is the next chapter in that story. And if history is any guide, it is one we’ll write together.