Google reveals Gemini Spark, Universal Cart

Google used its I/O 2026 keynote in Mountain View, California, on May 19 to introduce Gemini Spark and Universal Cart, extending its push into what executives called more “agentic” AI. The announcements were part of a broader product slate that also included AI features in Search, Android XR and Gemini Omni Flash, according to Google’s I/O roundup and keynote transcript. Google framed the updates around systems that can carry out tasks for users across its products, rather than only answering prompts. ### What, exactly, did Google say Gemini Spark does? Josh Woodward, vice president of Google Labs, Gemini app and AI Studio, said in a May 19 Google post that Gemini Spark is “an agent to help you get things done around the clock.” Google said the feature is part of a redesign of the Gemini app that also adds proactive “Daily Briefs” and a new interface. (blog.google) Google’s I/O collection page said the company is “unlocking agents and agentic experiences” across its products, naming “Information agents in Search, Gemini Spark and Daily Brief in the Gemini app” as examples. Sundar Pichai’s keynote transcript described I/O 2026 as the start of an “agentic Gemini era.” ### How does Universal Cart fit into that strategy? (blog.google) Google Shopping said on May 19 that Universal Cart is “an intelligent shopping cart” that works across merchants and Google services. The company said users can add products while browsing Search, chatting in Gemini, watching YouTube or reading Gmail, with the cart then tracking deals, price drops and stock changes in the background. (blog.google) Google said Universal Cart is powered by its Gemini models and is intended to become a central shopping hub across services. In a separate commerce update, Google said the cart can support checkout with Google Pay at major retailers and is tied to the company’s Universal Commerce Protocol. ### Where did Google place these products in the wider I/O lineup? (blog.google) The Keyword’s I/O recap listed Gemini Spark, Universal Cart and Gemini Omni among the event’s headline announcements. Google also highlighted Android XR, smart eyewear, Search agents and developer tools built around Gemini 3.5. Gemini Omni Flash, which Google introduced separately on May 19, is rolling out to Google AI Plus, Pro and Ultra subscribers globally through the Gemini app and Google Flow, according to the company. (blog.google) Google said the model can take images, audio, video and text as input to generate video, with broader developer and enterprise access planned in coming weeks. (blog.google) ### What did Google say about the shift from answers to actions? Google DeepMind executives Koray Kavukcuoglu and Jeff Dean wrote on May 19 that Gemini 3.5 combines “frontier intelligence with action” and is built for “complex, agentic workflows.” A separate I/O developer post said Google was “accelerating the shift from prompts to action” with Gemini 3.5 Flash. (blog.google) Sundar Pichai said in Google’s edited keynote transcript that the company was entering an “agentic Gemini era,” language that matched the framing in Google’s product and shopping posts. NDTV’s event recap also grouped Spark with AI-powered Search, Universal Cart, Android XR and Omni Flash in its coverage of the keynote. ### What comes next for users and developers? (blog.google) Google said Gemini Omni Flash is available now for paid Google AI subscribers and will expand to developers and enterprise customers in the coming weeks. Google’s I/O materials did not give a standalone launch date for broad Spark availability in the materials reviewed here, but the company presented both Spark and Universal Cart as live parts of its I/O 2026 product roadmap. (blog.google 1) (blog.google 2)