Woolworths to power Olive digital assistant with agentic AI

Woolworths Group will enhance its digital shopping assistant, Olive, using Google Cloud’s newly released agentic artificial intelligence (AI) platform, Gemini Enterprise for Customer Experience (CX).

Unveiled at the NRF 2026 conference in New York, the platform is designed to move retailers beyond static chatbots toward agentic systems capable of complex reasoning, maintaining context across channels, and executing tasks on behalf of the user.

Woolworths is the first Australian retailer to adopt the new platform. The retailer plans to use the technology to evolve Olive – originally launched as a customer service chatbot and digital helper – into what it calls an “intuitive partner”.

While the current version of Olive handles queries and assists with online orders, the upgrade will allow the assistant to use generative AI to anticipate customer needs. According to the retailer, the agentic Olive will be capable of planning meals based on specific tastes of shoppers and identifying specials that fit their budget.

“Google Cloud’s Gemini Enterprise for Customer Experience is a global game changer for retail,” said Amanda Bardwell, CEO and managing director of Woolworths Group, adding that the agentic AI capabilities in Olive will be a “practical innovation that’s all about us doing the heavy lifting for you, making shopping that little bit easier to give you time back in your day.”

Besides Woolworths, US grocery giant Kroger is also using the same platform to personalise meal planning for customers, while Lowe’s said it is using it to provide “guidance customised to a customer’s home improvement project and where they live, bringing new levels of confidence to every decision.”

Google Cloud described the Gemini Enterprise for CX as a proactive digital concierge that allows businesses to deploy agents that use complex reasoning to understand intent and execute multi-step tasks on behalf of a customer, taking into account their preferences and consent.

The system also supports multimodal interactions, allowing the AI to process voice, image, and video inputs without requiring the customer to type search queries.

“With Gemini Enterprise for Customer Experience, we are combining the best of Google Cloud’s AI and infrastructure with a business’s own institutional intelligence to power a truly agentic commerce journey,” said Darshan Kantak, vice-president of applied AI at Google Cloud. “By bridging the gap between sales and service, businesses can deliver premium, personalised experiences, from initial discovery to post-purchase support.”

Google Cloud also announced the launch of the Customer Experience Agent Studio, a development environment that allows retailers to build and deploy multimodal agents using a drag-and-drop interface. It claims the tool allows businesses to turn existing chat transcripts and documents into functioning AI agents in a matter of days.

Following the launch of the Agent Payments Protocol (A2P) last year, Google Cloud has also developed the Universal Commerce Protocol (UCP), a new open standard for agentic commerce that works across the entire shopping journey.

UCP establishes a common language for agents and systems to operate together across consumer surfaces, businesses, and payment providers. Google Cloud said the protocol was built to work across industries and is compatible with existing industry protocols, including A2P, Agent2Agent (A2A), and the Model Context Protocol (MCP).