Online shopping has been defined for decades by consumers’ long-winded search-browse-and-buy pursuits across the web, keeping buyers close to brands they value and fully embedded in the journey. But now, Google’s newly unveiled Universal Cart, built on AI models running seamlessly in the background of every web-based surface, has the power to dramatically shift the dynamic.

As the dominant consumer internet software company that powers around 90% of global web searches, Google boasts it has “the world’s most comprehensive catalog” of over 60 billion product listings. Already, this shopping experience is powered by Google’s advanced AI and Shopping Graph, its machine learning-powered database of all the product information on the internet.

Google’s new Universal Cart, announced at its I/O 2026 conference on Tuesday, capitalizes on this market share and represents the most significant shift in e-commerce from search-based browsing to what the industry calls “agentic commerce” to date. Billed as an “intelligent shopping cart and your new agentic hub for shopping”, it stands as the core of the tech giant’s AI shopping hub, allowing internet users to add products to a single cross-platform cart throughout its suite of applications, including Google Search, Gemini, YouTube, and Gmail.

Google plans to start rolling out the Universal Cart across Google Search and the Gemini app in the United States in summer 2026, with YouTube and Gmail to follow. The new checkout features will first go live across brands, including Nike, Sephora, Target, Ulta Beauty, Walmart, and Wayfair, as well as some Shopify merchants, such as Fenty and Steve Madden.

Google also announced a new 24/7 AI personal assistant, Gemini Spark, on Tuesday, which it plans to integrate with the Universal Cart. This paves the way for Spark to eventually handle the shopping journey end-to-end and make purchases on consumers’ behalf — the true definition of an AI shopping agent.

In a practical sense, this means there is no wasted moment in the shopping experience on a Google-based platform. The moment a product is added to buyers’ baskets, the agentic AI comes into play, continuously searching for the best deals, comparing compatibility between selected items, creating price alerts, and checking stock of products spanning Google’s entire search index. In other words, agentic shopping will leave no stone unturned.

It’s what experts say represents a shift from Google designing AI shopping for consumers to use as a discovery tool, into a “decision-making” agent that could minimize the need to engage directly with brands — and dramatically shift consumer shopping behavior.

“Universal Cart normalizes the expectation of agent-assisted decision-making at a mass consumer level,” says Holly Enneking, VP of marketing at Markup AI, an AI content software company. “This sets up a future where B2B buyers arrive at a sales conversation already filtered, scored, and pre-decided by AI, not by your storytelling, or your brand equity.”

What this means for brands

Introducing agent-assigned decision-making as the new middleman between brands and consumers in the purchasing journey has some technologists predicting that brands’ DTC touchpoints could be taken out of the loop.

“In two to three years, the purchase journey as fashion brands have understood it will largely no longer exist,” says Max Sinclair, CEO of Azoma AI, which helps brands with their AI search strategies. “The shopper will set an instruction, an agent will execute it on their behalf, and the brand will find out it was selected, only when the order lands.”

This means that when a consumer makes a purchase in Google’s new cart after searching, checking emails, or watching YouTube, brands have fewer opportunities to influence the decision at the point of sale than they previously did through creative, merchandising, and last-minute upsell. Additionally, a major caveat is whether the agent even surfaced the brand in the first place.

Ever since Google and ChatGPT introduced dedicated AI shopping features this time last year, brands and marketers have been reshaping their online brand visibility so that products can be surfaced by both traditional search engine algorithms (via SEO) and AI models, via the emerging practice of “AIO” (AI optimization). Now, Google’s Universal Cart puts an AI agent front and center, making the first round of decisions, rather than a consumer simply asking a chatbot a question and perhaps consulting traditional search at the same time.

“With Universal Cart alongside Google’s Agent Payments Protocol, consumers are setting parameters for agents to buy on their behalf, which means a brand is either in the agent profile, or it’s not in consideration, full stop,” says Enneking. “That’s a massive barrier to break through. It’s not just about affinity anymore; it’s about whether your brand relationship is strong enough to get encoded into how someone’s AI agent shops for them.” As it stands, Enneking doesn’t think many brands are ready to confront or adapt to this shift.

According to Sinclair, the fields that power agentic reasoning, including product details like fit, materials, care, and compatibility, are simultaneously “the most ignored real estate” in e-commerce. This compounds with additional factors outside a brand’s control, such as online consumer product reviews and discussions in online forums like Reddit, which have seen a resurgence in utility in recent years and are known to aid AI’s processing of information.

“Brands auditing those attributes now are quietly setting themselves up to be the ones Universal Cart recommends. The ones ignoring them are about to find out how expensive that decision was,” Sinclair says. Beyond product metadata, experts say Google’s new AI shopping feature transforms brands’ and retailers’ inventory accuracy into a consumer-facing metric.

“The moment a shopper can add a jacket to a cart from inside a Google search result, your inventory number becomes a purchase promise displayed to millions of users,” says Ben Hussey, co-CEO of inventory management platform Katana Cloud. “Now, a 97% accurate inventory figure that was acceptable when it lived inside your Shopify admin becomes a source of abandoned carts and poor Google Merchant ratings when surfaced in a universal checkout flow.”

This means Google’s new rollout will require merchants to drastically improve their backend operations from an inventory standpoint to be able to update them much faster, in real-time, closing the gap between operational failure and customer-facing failure. The closest example to what this will look like is currently the impact of product virality on social media.

“A product goes viral on TikTok or Instagram, demand spikes 300–400% in 48 hours, and brands are making restock and transfer decisions with stale data,” says Hussey. “This is directly relevant to how Universal Cart will behave: when an item trends in Google Shopping, demand can become non-linear very quickly.”

According to Hussey, the most common cause of merchants’ fill rate (orders filled completely and on schedule) underperformance isn’t demand forecasting but inaccurate routing of inventory that exists in the system to the wrong channel or location. “That’s a synchronization problem, and it’s exactly the problem Universal Cart will make more visible, more quickly,” he adds.

Power shifts to Google

Now that AI is reshaping the way consumers behave online, Big Tech’s new products and features are rarely a passing fad. Where OpenAI seems to be retreating from hosting as much of the shopping journey via its rollout of merchant experiences within ChatGPT, like Asos’s new AI stylist launched this week, Google’s Universal Cart is a major push towards owning the entire shopping journey, end-to-end.

Where typically, the initial idea of excessive data and privacy infringement by Big Tech companies worries users, over time, consumers often find themselves leaning into these features due to the convenience of the end result, or the act of withdrawing personal data being too complicated.

It’s something that’s already showing up in the data on the AI Overviews that appear at the top of Google search. AI Overviews led to a 58% lower average clickthrough rate for top-ranking pages as of May 2026, up from 34.5% just eight months ago, according to new research from the AI-powered marketing platform Ahrefs. It’s an uptick that implies consumer behavior is changing faster than most are perhaps aware, and a reminder that Google’s software updates hold norm-altering power.

“These AI agents and the LLMs that power them are probabilistic [i.e. impossible to predict] statistical models that are having private one-on-one conversations,” says Matt Maher, founder and CEO of fashion-tech consultancy M7 Innovations. “This is a data black box that no retailer will ever have access to — it’s about as uncomfortable as a brand can get to knowing they’re losing visibility on the consumer journey.”

Maher’s advice is that brands focus on what they can control — their data hygiene and site infrastructure — to play into this probability. “These agents couldn’t care less about the brand equity your brand has built over the years; they’re simply looking for the answers,” Maher says. “So if your data doesn’t have those available in an easily accessible way, the agents will go to your competitor or whoever does have it.”

As with all AI-related innovations, there’s a constant pendulum of worry between where consumer convenience ends and data and privacy concerns begin. The Universal Cart pulls a consumer’s data, history, and habits across Search, Gemini, YouTube, and Gmail, which holds the potential to create a more detailed portrait of a shopper than any that has existed before.

The system has also been built on top of Google’s pre-existing Google Wallet service and therefore instantly understands consumers’ payment-method perks, loyalty information, and merchants’ offers to buyers. Whoever hosts the final transaction in the shopping journey dictates the user’s experience of the critical point of sale. Google’s new Universal Cart additionally gives consumers the option to “Buy Now” or re-route to the merchant’s site to complete their purchase. But the convenience of the Buy Now option, coupled with hyper-personalization, has Sinclair predicting that “most people will take the trade” of syncing their Google Suite to the new cart, which could quickly drive up adoption.

This means the Universal Cart will give Google more data than ever about who buys what, in which color, and at what price point — data that typically drives brands’ next-season buying decisions and design iteration. Google clearly states in the terms and conditions of its Universal Commerce Protocol (the precursor to this new Universal Cart update) that brands “remain the Merchant of Record, maintaining full control over customer relationships and data” — but is yet to specify whether this is still the case with the new Universal Cart. Sinclair warns that this could have brands grappling with essential data access.

“Brands operating primarily through Universal Cart will increasingly be making those calls on incomplete information, feeding Google’s model, while starving their own,” Sinclair says.

Brand world-building still matters

As brands reinvent themselves on the commerce end of their consumer chain, experts say AI has yet to crack the crucial notions of brand loyalty and customer retention, which remain within brands’ power and will become more important than ever.

Brands that come out on top will have their eyes set on improving their data visibility while continuing to build emotional consumer touchpoints with their brand that tech platforms like Amazon and Google categorically cannot.

“The reason a customer buys a second Ganni piece, queues for a Rowing Blazers drop, or participates in a Depop resale community isn’t transactional,” says Hussey, stressing that this demand is social and identity-driven. “Community lives in the spaces around the purchase: the email series, the Discord, the try-on video, the founder’s Substack. Those touchpoints don’t route through Google’s checkout,” says Hussey.

“Scarcity, ritual, narrative, co-creation: these are the tools of community building that remain brand-owned, regardless of where the transaction occurs.”