AI and the future of online shopping: an imminent change

Table of Contents

Artificial intelligence is transforming online shopping

  • Online shopping is undergoing its biggest change since the invention of the search engine, driven by conversational interfaces.
  • Platforms like ChatGPT are beginning to concentrate discovery and purchase into a single conversation.
  • A Morgan Stanley survey suggests that nearly 1 in 4 Americans already used AI to shop in the last month.
  • Advertising inside AI opens up a tension: monetization vs. consumer trust.
  • Many brands admit they’re late: they say they value the customer experience, but acknowledge they’re lagging in AI.

From the search engine to the conversation
– Before (search-engine era): “I search → click a blue link → retailer website → filters/categories → checkout”.
– Now (conversational era): “I describe what I need → the AI asks/adjusts → recommends → (increasingly) lets me buy without leaving the chat”.
– What fundamentally changes: the “entry point” stops being a page and becomes an answer; that can shift visibility, brand control, and trust toward the platform that drives the conversation.

How AI is changing online shopping

For 25 years, e-commerce relied on an almost invariable ritual: search for a product, open a “blue” link, jump to a retailer’s website, and complete a checkout with linear steps. That pattern—which defined the search-engine era—begins to give way to a different interface: conversation.

Dan Gardner, founder and CEO of Code and Theory (the creative agency behind NFL.com and digital experiences for brands like Microsoft, Google, Time, and Vogue), sums it up as a change in how the internet is used, not just in tools. In his view, people are “chatting with websites” and finding a form of personal experience “completely different” from that of the last decades.

The promise of AI in shopping is not only to recommend products: it is to compress the entire journey. Instead of browsing categories, filters, and pages, the user describes what they need and the platform responds, asks, adjusts, and guides. Gardner describes it as a “non-linear” flow that adapts to “the way you think,” and that feels “more human” even though it is technologically more advanced.

The tipping point, according to this view, is that AI stops being a help layer and becomes the place where the purchase happens. That reshuffles power: whoever controls the conversation can end up mediating the relationship between brands and consumers, redefining what is seen, what is recommended, and what is bought.

Purchase process

conversational
1) Intent: the user explains need, budget, urgency, and context (“as a gift,” “for running,” “delivery tomorrow”).
2) Clarification: the AI asks 1–3 questions to reduce ambiguity (size, compatibility, preferences).
3) Selection: it proposes options and justifies them (why they fit what was requested).
4) Quick validation: the user adjusts (“cheaper,” “another brand,” “no subscription”).
5) Close: purchase within the chat or minimal jump to payment/confirmation.
Practical checkpoint for brands: if your catalog, pricing, stock, shipping, and returns aren’t well structured and accessible, the conversation breaks right at step 3–5.

Morgan Stanley survey: the use of AI in shopping

Terminology note: in the text, “AI” and “IA” are used as equivalents (artificial intelligence), as they appear in the original cited material.

Adoption is no longer a lab hypothesis. A recent Morgan Stanley survey found that approximately one in four Americans used AI to make a purchase in the last month. The figure, cited as a sign of early traction, suggests that behavior is crossing over from “trying out of curiosity” to becoming integrated into consumer habits.

The data matters for two reasons. First, because it indicates that AI already participates in transactional decisions, not just informational ones. Second, because it anticipates a snowball effect: if the user perceives that the conversation reduces friction — fewer tabs, fewer manual comparisons, fewer steps — the incentive to repeat grows.

Gardner argues that this percentage “will only increase.” His argument is supported by product direction: conversational platforms are expanding capabilities so the user doesn’t have to leave the chat. In that context, AI functions as a “front door” to commerce: the place where the purchase is initiated, evaluated, and closed.

The takeaway for the market is clear: if a meaningful share of consumers already buys with AI, the channel stops being marginal. For brands, the risk isn’t only losing web traffic; it’s losing control over how their offering is presented when the dominant interface is no longer a page, but a response generated in real time.

AI in the purchase process
– Cited data: “approximately 1 in 4 Americans” used AI to make a purchase in the last month (Morgan Stanley survey, according to the summarized note).
– What it suggests (without overinterpreting): it’s a sign of early adoption with transactional behavior, not just search.
– How to read it in context: a survey captures a moment and a sample; eventhus, the key point here is the direction of change: AI is already entering the act of purchase.

Platforms like ChatGPT and their impact on the shopping experience

Gardner points to ChatGPT as “ground zero” of the transformation. The key change is that, where previously the consumer had to leave the search engine and browse sites, now they can discover and buy within the same conversation. ChatGPT, according to the account, recently added the ability to transact directly on the platform: the user can go from “I want X” to “I buy it” without leaving the chat.

That integration alters the shopping experience in two dimensions. The first is continuity: the user doesn’t “bounce” between pages, but maintains context. The second is dynamic personalization: the conversation makes it possible to refine preferences in real time (“cheaper,” “faster,” “for a gift,” “for such-and-such use”), without forcing the user to learn the navigation logic of each store.

But the impact isn’t limited to the consumer. For brands, the conversational interface can become an intermediary that decides which options appear first and how they are explained. In the classic model, the retailer controlled its digital storefront; in the conversational model, part of that storefront shifts to the AI platform.

Gardner describes it as a “one-stop” and “non-linear” experience, aligned with the human way of thinking. In practice, that can raise the bar: if buying becomes as fluid as a chat, any friction—long forms, redundant steps, confusing information—becomes more visible and less tolerable.

Control of the customer relationship
– In favor (for the user): less friction, more continuity of context, guided comparison, and real-time adjustments.
– In favor (for the brand): the possibility of showing up at the “moment of intent” with more relevant messages and less drop-off due to navigation.
– Against (structural risk): more intermediation: the platform can decide the order, framing, and visibility of options.
– Sensitive point: if the user “trusts the answer” more than the store, part of the value (and control of the relationship) shifts from the brand’s site to the assistant.

AI applications in sports: personalization and promotions

Sports offers an advanced laboratory for what AI can do when combined with behavioral data and an emotional relationship with the brand. Gardner explains that his firm worked on applications for leagues and teams—including the NFL and the Philadelphia Flyers of the NHL—where AI already tailors experiences on an individual basis.

What does that translate intopersonalization? In that the content, promotions, and even the order of the feed can vary depending on each fan. The logic is fed by signals such as purchase history, team loyalties, and interaction patterns. Instead of the same front page for everyone, the app “steers” the type of content and the format to try to ensure “the best possible experience.”

The commercial implication is direct: if the platform understands what a fan consumes and buys, it can present more relevant offers at the right time. And, at the same time, it can strengthen the bond: not just sell, but keep the user engaged with content that fits their sports identity.

Gardner insists that this is not a future possibility, but something that “is happening.” In the context of e-commerce, the parallel is obvious: AI will not only recommend products, but will reorganize storefronts, messages, and interaction sequences for each person, with the goal of maximizing perceived utility and conversion.

Personalization in Three Layers
Personalization in 3 layers (from sports to e-commerce):
– Signals: purchase history, affinity (team/brand), behavior (clicks, time, frequency), context (event/season).
– Decision: what to show first (feed/storefront order), what format to use (video, text, featured), what incentive to offer (promo, bundle, shipping).
– Expected result: more perceived relevance, more interaction, higher conversion and retention (without relying only on “more ads”).

Challenges of advertising in an AI-driven world

As AI concentrates attention, the uncomfortable question arises: how do you monetize without eroding trust? OpenAI announced plans to introduce ads in its free tier. For Gardner, it was inevitable: “where there is attention, there is advertising.” If millions of users spend time in a conversational interface, the advertising incentive comes with them.

The problem is that advertising in an answer environment can feel less visible than a banner or a traditional sponsored result. OpenAI committed to clearly labeling sponsored content, in a logic similar to the separation between paid and organic results on Google. Gardner believes the distinction matters, but it does not ensure it solves the underlying dilemma.

When asked whether consumers will really be able to know that AI recommendations are not shaped by advertisers’ money, his answer is blunt: “The reality is they won’t know, and that’s going to be a danger.” In other words, even with labels, the perception of bias can take hold.

His counterweight is competition. With alternatives like Google Gemini, Claude, and “dozens” of platforms,

Gardner believes consumer choice will push to maintain integrity: if an AI is perceived as too influenced by ads, it will pay a reputational cost and users will migrate. The market, he suggests, could punish opacity, but the trust risk will remain structural.

Tension In search engines (known model) In conversational assistants (what changes) Risk to trust
Organic vs. sponsored Relatively clear visual separation (blocks/positions) The recommendation may “feel” like a single answer The user may not perceive where organic ends and paid begins
Labeling “Ad/Sponsored” is usually in plain view Labels must coexist with generated text and a conversational tone If labeling is subtle, it is interpreted as manipulation
Explainability The user can open multiple sources The AI summarizes and decides what to show Less ability for the user to audit for themselves
Incentives The search engine monetizes clicks/impressions The assistant monetizes attention and, potentially, transaction Suspicion of bias in “default” recommendations

How companies are preparing for the shift in e-commerce

If consumers move toward conversational purchasing, the question is whether companies are ready to follow them. Gardner argues that the gap is widening and that most brands are behind. He cites a Code & Theory survey conducted with The Wall Street Journal: 94% of executives agree that a great customer experience is vital to their company’s future, but 93% admit they are already lagging in delivering it. And 75% say they are falling behind specifically in AI.

The contradiction, according to Gardner, stems from how AI is being understood within organizations. Many treat it as an internal efficiency tool—cutting costs, automating tasks—instead of using it as a customer-oriented creative lever. By contrast, he argues, tech companies are using it to reimagine the experience.

His advice for “legacy” brands that could suffer AI-driven digital disruption is to strengthen what AI cannot replicate: brand meaning. If AI tends to homogenize content and responses, brands with a clear, resonant identity will bethat people deliberately seek out. Gardner illustrates this with a comparison: the reason someone buys an expensive handbag instead of a free tote bag is brand resonance.

The final risk is strategic: brands that don’t build AI-driven customer experiences don’t just fall behind; they cede ground to platforms that could end up mediating “every interaction” between them and their consumers.

Operational Readiness for AI
Quick readiness signals (so you don’t get stuck at “yes, it’s important”):
– [ ] Product data ready for conversation: catalog, attributes, compatibility, and consistent FAQs.
– [ ] Reliable pricing/stock/shipping: up-to-date availability and timelines (if not, the AI promises and operations suffer).
– [ ] Clear policies: returns, warranties, and support explainable in 2–3 sentences.
– [ ] Minimum integrations: payments, CRM/tickets, inventory, or ERP (depending on the business).
– [ ] Human scaling design: when the bot should hand off to an agent and how context is preserved.
– [ ] Experience metrics: first-contact resolution, abandonment, satisfaction, and escalation reasons.

Implications of AI in e-commerce

The New Era of Customer Experience

AI is pushing e-commerce toward an experience centered on dialogue, context, and continuity. In the model described by Gardner, the conversational interface is not an accessory: it is the new “counter” where what to buy and how is decided. Shopping stops being a rigid sequence of pages and becomes an interaction that asks, interprets, and proposes.

“The way we’re starting to use the internet is fundamentally changing.”
Dan Gardner, founder and CEO of Code and Theory

That “new era” also redefines expectations. If the user can resolve in a conversation what previously required multiple clicks, the standard of convenience rises for all channels. Customer experience comes to be measured by flow: fewer jumps, less friction, more relevance.

Challenges and Opportunities for Brands

The opportunity is clear: personalization at scale, more useful recommendations, and shorter purchase journeys. But the challenge is just as big: losing control of the primary touchpoint if the conversation happens on someone else’s platform.

Advertising adds another layer of tension. If AI is funded by ads, trust becomes a fragile asset. Gardner warns that the consumer won’t always be able to tell whether a recommendation is influenced by money.In that scenario, the platform’s reputation and the clarity of labeling become decisive, but not infallible.

“The reality is they won’t know, and that is going to be a danger.”
Dan Gardner

Competition between platforms appears as a disciplining mechanism: if there are alternatives, the user can punish experiences perceived as biased. For brands, this means operating in an ecosystem where visibility may depend on changing rules and on intermediaries with their own incentives.

The Importance of Authenticity in the Digital Age

When AI “washes” content down to a common denominator, authenticity becomes a differentiator. Gardner argues that, in an environment where many answers look alike, meaningful brands—with a recognizable and consistent proposition—will be the ones that generate preference.

Authenticity, in this framework, is not a slogan: it is a way to resist commoditization. If AI makes it easier to compare and substitute, the brand needs something more than price or availability to remain in the consumer’s mind. Resonance—that which makes someone choose an option “because it’s that one”—becomes a competitive advantage.

“When everything gets washed down into its lowest common denominator, people will actually want some resonance with something that has meaning.”
Dan Gardner

Key implications and priorities
Quick map of implications (to prioritize):
– Customer: less friction and more guidance; expectations for immediacy and clarity rise.
– Brand: differentiation through meaning/identity; risk of “commoditization” if everything is answered the same way.
– Channel: the entry point may shift from the site/app to the conversation; the acquisition logic changes.
– Trust: sponsorship labeling, perceived bias, and the user’s ability to verify influence adoption.

Transform the customer experience with Suricata Cx

The revolution in customer service for ISPs and telecommunications

In telecommunications and ISPs, the customer experience is often defined by volume: many repetitive inquiries, demand spikes, multiple channels, and pressure to resolve quickly. Suricata Cx positions itself as an AI-powered omnichannel platform, designed specifically for telecom and ISP operations in the Americas and Spain, combining automation, conversational AI, and flows with human oversight.

This is not a generic chatbot: the focus is on operating on real industry processes (support, sales, payments, and recovery), with

operational integrations and conversation traceability across channels such as WhatsApp, webchat, social networks, and IVR/voice.

Tangible benefits of implementing artificial intelligence in your operation

The proposal is based on applying AI where it is predictable and automatable, keeping people in control when judgment is needed. In practice, this aims to reduce costs per interaction and response times, improve first-contact resolution, and sustain quality without proportionally scaling the team.

In sales, automation can help qualify leads and guide the user through the purchase journey in conversational channels. In payments and collections, conversational flows and reminders are proposed, with service reactivation after payment, supported by a gateway specialized in the sector.

How Suricata Cx adapts to the specific needs of your business

Suricata Cx is described as “API-first” and geared toward integrations with billing systems and ISP ERPs, with real-time data synchronization (customers, debts, services, tickets). It also incorporates a “human-in-the-loop” model: the bot can escalate to agents, pause to receive human input, and resume, with control and auditing for supervisors.

At a time when AI is rewriting the way people buy and engage with digital services, the advantage is not only automation: it is designing conversational experiences that maintain context, reduce friction, and reinforce trust—especially in industries where service is a central part of the brand.

AI and the future of online shopping: an imminent change reminds us that conversation will be the new “checkout” and that trust is earned by reducing friction without losing control. From Suricata Cx, that same logic translates into omnichannel experiences for telecom and ISPs where AI automates what is predictable and escalates to people when judgment is needed, protecting the customer relationship.

At Suricata Cx, these kinds of changes are analyzed from the perspective of omnichannel operations and “human-in-the-loop” flows in telecom and ISPs: what to automate, where to maintain human control, and how to sustain traceability and integrations so that the conversational experience is consistent.

From pilot to operational scale
Typical implementation (from pilot to scale) in telecom/ISP:
1) Discovery (1–2 weeks): priority use cases (support, sales, payments), channels, and target metrics.
2) Integration (2–6 weeks): connection with billing/CRM/tickets; definition of minimum data (customer, debt, service status).
3) Controlled pilot (2–4 weeks): one channel and 1–2flows; review of real conversations and escalation points.
4) Scaling: more flows/channels; operational training (supervision, auditing, continuous improvement) and clear “human-in-the-loop” rules.
Operational checkpoint: if there is no traceability (what was said, what was resolved, why it was escalated), it is difficult to improve without affecting the experience.

This text reflects publicly available information at the time of writing and retains figures and claims attributed to their sources. Purchasing functions on conversational platforms and advertising plans can change quickly as products evolve. If you are going to make channel or investment decisions, cross-check against more recent public information and your own data, as there may be uncertainties or updates.