Tesco, the UK's largest supermarket chain, is moving beyond predictive analytics and building AI agents that can take action, from planning meals to automatically filling shopping baskets. The retailer, which accounts for almost 30% of the UK grocery market, is developing the technology through partnerships with Adobe and French AI developer Mistral. The shift signals a broader move toward autonomous retail, where AI does not just suggest what to buy but actively participates in the buying process.
The work is still early. No concrete results from the project are available yet, and the AI assistant is currently in pilot, being tested by staff before any customer rollout. But the direction is clear: Tesco wants AI that acts, not just advises.
From Loyalty Data to Agentic AI
Tesco has spent 30 years collecting purchase data via Clubcard, its loyalty program. That data now covers eight out of 10 UK households, giving the retailer a picture of British shopping habits that few competitors can match. For years, Tesco used that information for predictive analytics and recommendation engines, suggesting products based on past behavior. Now it is pushing further.
Agentic AI refers to systems that can take actions, not just provide insights. In Tesco's case, that means AI that can plan a week of meals, adjust for dietary needs, suggest ways to use leftovers, and build a shopping basket automatically. The assistant will live inside the Clubcard app, turning a loyalty tool into an active shopping companion.
The technology is built around Adobe's Firefly Foundry, a genAI platform designed for personalized, branded creative materials at scale. Tesco's data and AI teams work alongside Adobe engineers in the Tesco x Adobe Innovation Lab, a strategic partnership announced in April 2026. The lab is the platform for co-developing bespoke agentic solutions.
The Assistant in Action
The AI assistant provides meal planning advice tailored to individual preferences and dietary needs. It also offers advice on reducing food waste, suggesting how to use leftovers in new meals. Most notably, the assistant is reported to automatically build shopping baskets based on preference and purchase history.
That last feature is the most significant step. A recommendation engine tells you what to buy. An agent adds it to your cart. The difference may seem small, but it changes the relationship between retailer, customer, and machine.
The assistant is being tested by staff first. That internal pilot allows Tesco to refine the system before exposing it to millions of Clubcard users. The company has not announced a customer rollout date.
Mistral Partnership and the Trust Barrier
Beyond customer-facing AI, Tesco has an agreement with Mistral for internal operations. The collaboration includes content development, document drafting, and data analysis. It also aims to develop new ways for colleagues to access information.
A Tesco spokesperson said the company is "developing new ways for colleagues to easily access information which they can use to help customers." That quote captures the dual nature of the project: efficiency inside the business, better service outside it.
Stay ahead of the AI curve
The most important updates, news, and content — delivered weekly.
No spam. Unsubscribe anytime.
The Mistral work is less visible than the Adobe partnership, but it matters. If staff can find answers faster and draft documents more quickly, they have more time for customers. The internal tools are a quieter but real part of Tesco's AI strategy.
The real challenge for autonomous retail is earning customer trust. If customers let AI make small decisions, the bigger question is how long before they let it make purchases. That is not a technical problem. It is a human one.
Tesco has an advantage here. Thirty years of Clubcard data means the AI has a deep understanding of individual shopping habits. That personalization could make the assistant feel helpful rather than intrusive. But data alone does not create trust. Customers need to see that the AI works, that it respects their preferences, and that it does not push products they do not want.
The project is new, so no concrete results are available yet. That means the trust question remains open. Tesco is testing internally, refining the assistant, and preparing for a customer launch that has not yet been scheduled.
The Bigger Prize and What It Means for Retail
The article, published on Aug 12, 2026, by Forbes contributor Bernard Marr, argues that Tesco's experiments point toward a future where AI agents shop for us. The bigger prize is a future where AI agents handle the entire shopping trip, from meal planning to checkout.
Tesco isn't there yet, but current experiments point in that direction. Each useful interaction gives customers another reason to hand over a little more control. Meal planning is a small step. Automatic basket building is a bigger one. Full purchase automation would be the largest.
The progression is logical. Once an AI knows your preferences, dietary needs, and budget, it can make many decisions for you. The question is how much control customers are willing to give up.
The article sends a clear message that retailers are moving beyond traditional predictive AI towards agentic models. Tesco is not alone in exploring this space, but its scale makes it a notable case. With almost 30% of the UK grocery market and loyalty data covering most households, Tesco's experiments could shape how the entire industry approaches AI.
The shift from influencing purchases to actively participating in buying is significant. It changes what AI does in retail, from a passive tool that generates recommendations to an active agent that completes tasks. That shift carries both opportunity and risk.
For now, Tesco is moving carefully. Staff are testing the assistant. The Mistral tools are being developed for internal use. The Adobe lab is building new capabilities. No customer rollout has happened, and no results have been published.
But the direction is set. Tesco is building AI that does not just understand shoppers. It is building AI that shops for them. The next few years will show whether customers are ready to hand over the basket.