Dept. of Mysteries
Google Releases A2UI 1.0, a Wondrous Decree for Agents That Answer with Screens
Google releases A2UI 1.0, the first stable version of its open protocol for generative UI, letting AI agents answer with interactive screens instead of text.
Wednesday, 11 November 2026 · 2 min read
Behold, an answer to the wall of text. Google has released A2UI 1.0, the first stable version of its open protocol for generative UI, which is to say interfaces that an AI agent assembles on the spot rather than pages built in advance. The protocol lets AI agents answer with interactive screens, such as a form to confirm or a chart, in place of the lengthy prose that so often greets a user today. The arrangement is curiously courteous. The agent describes what to show, and every app draws the UI with its own native components, so agents use a company's design system and nothing else. A2UI binds to data and child components on the server, so sensitive values never reach the model, and the app streams in the UI as the agent works. Developers may wield it at once, with open-source renderer SDKs for the web, Android, and iOS and agent SDKs for Python, TypeScript, and Dart. Renderers draw A2UI screens with React, Angular, and Lit on the web; Flutter on mobile, web, and desktop; SwiftUI on iOS; and Jetpack Compose on Android. The Jetpack Compose renderer is currently v0.9.1, with v1.0 support coming soon. Version 1.0 also adds bidirectional function calls, so a flight or reservation picker can check availability on the spot, and the protocol adds support for accessibility. The annals already record early adopters. Companies such as Atlassian, Alation, Bloomberg, and ByteDance are building with A2UI. Atlassian is bringing it to Jira, where an agent needing a missing detail or decision asks with an interactive card in the work item, while ByteDance released an A2UI renderer for Lynx, its cross-platform framework, in June 2026. Within Google itself, Gemini Notebook uses A2UI cards to ask how a slide deck should be shaped before it is generated, and Gemini supports A2UI 0.9.1 today for organizations that register their own agents. The web and Flutter renderers also work inside MCP Apps, the apps that a third-party AI assistant can host, so A2UI screens can appear in AI assistants that support MCP Apps, such as Claude and Gemini. Google says several other teams, including some behind its largest products, are building features with A2UI that have not launched yet.
Source: developers.googleblog.com