AI shopping: Shopify and Google introduce standard for purchases via AI agents
The use of AI for product discovery, comparison, and purchasing is gaining significant momentum, and Shopify is strategically capitalizing on this trend. Together with Google, the company has introduced the Universal Commerce Protocol (UCP): an open standard that allows AI agents to handle purchases directly with brands. The protocol operates within AI environments such as Google AI Mode, Gemini, Microsoft Copilot, and ChatGPT. Shopify announced this development during NRF 2026 and via a press release.
Thanks to the UCP, AI agents can automatically apply discounts, loyalty programs, subscriptions, and sales conditions during a chat conversation, allowing consumers to complete a purchase entirely within the AI environment. American brands Gymshark, Everlane, and Keen are already utilizing the technology.
Because it is an open standard, all brands can sell their products via AI—even if they do not utilize Shopify technology for their online store. In this instance, Shopify functions as the technical infrastructure behind the transactions via the new ‘Agentic plan’.
With this launch, the companies are focusing on scaling agentic commerce, a new technical buzzword in the e-commerce sector and industry jargon that is increasingly surfacing in the fashion sector as well.
Agentic AI and Agentic commerce: what is it?
Agentic AI represents the next step in the evolution of artificial intelligence. Whereas we are currently accustomed to AI primarily answering questions or generating text, an AI agent actually goes to work for you. You do not give an agent instructions for every single step; instead, you provide it with an end goal. The agent then determines for itself which steps are necessary: it creates a plan, retrieves the correct information, utilizes external software, and adjusts its course if something goes wrong. It is no longer a chatbot, but a digital assistant that performs complex tasks independently from start to finish. Major technology companies such as Google, Amazon, Meta, and OpenAI are investing heavily in this technology.
Agentic commerce is an application of agentic AI within e-commerce. Here, AI agents play an active role in the purchasing process: they can recommend products, apply discounts, finalize payment processes, and execute other transaction steps independently, enabling consumers to make a purchase without manually going through all the steps themselves.
AI shopping traffic is on the rise: these are the opportunities and challenges
The use of AI for product orientation and online shopping is clearly on the rise. Although direct traffic to online stores is decreasing, the share of visitors arriving via AI tools is growing strongly. According to Adobe, traffic from AI sources to retail websites in the United States increased by a staggering 4,700 percent in one year, as reported in October.
Visitors often arrive with a high sense of trust: the AI agent has directed them to a specific website, suggesting that this is the right choice. This translates into higher expectations and a stronger purchase intent. Those who land on a site via generative AI stay an average of 32 percent longer and demonstrate a 27 percent lower bounce rate, according to Adobe.
Salesforce calculated that approximately 20 percent of total online retail sales during the past holiday season were influenced by AI and AI agents. AI was frequently deployed, particularly for personalized recommendations and advice.
The recently published ChannelEngine Marketplace Shopping Behavior Report 2026 confirms this trend. The survey of 4,500 consumers in the US, the UK, France, Germany, and the Netherlands shows that 58 percent use AI tools for product research and 37 percent have actually initiated a purchasing journey via an AI assistant. At the same time, the research also reveals a clear threshold: the 'trust gap'. Only 17 percent of consumers feel comfortable enough to let AI complete a purchase entirely. Although orientation is taking place en masse via AI, for the time being, the consumer prefers to keep the reins in their own hands for the actual transaction.
Sources:
-‘Shopify unveils agentic commerce with native buying in AI at NRF’, press release 12 January 2026.
-AI tools were used for writing assistance.
-Salesforce insights ‘Holiday Season Rakes in Record $1.29T for Retailers, Salesforce Data Shows’, 8 January 2026
-ChannelEngine ‘Only 17% of Consumers Trust AI Enough to Complete a Purchase, Global Study Finds’, 7 January 2026
OR CONTINUE WITH