論文
Generative UIに関する学術研究と技術論文
事例
Generative UIを活用した実際のプロダクトとアプリケーション
動画
Generative UIのチュートリアル、デモ、講演
リソース
ブログ、ドキュメント、SDK、OSSプロジェクト
議論
コミュニティフォーラムと注目スレッド
最新リソース
ag-ui-protocol/ag-ui
The official AG-UI protocol repository, positioned as the agent-user interaction layer for bringing agents into frontend applications. It is especially relevant because it sits one level below app-specific demos: it defines the transport and event model that lets agent backends stream messages, tool activity, and UI updates into real products across clients.
Macaron-A2UI: A Model for Generative UI in Personal Agents
Fresh 2026 arXiv paper focused on runtime Generative UI for personal agents. The authors argue that plain chat becomes a bottleneck for complex agent tasks, introduce a large-scale Generative UI corpus plus A2UI-Bench, and train models that generate natural language together with lightweight executable UI actions for confirmation, preference refinement, and multi-goal coordination.
Google Stitch: Real-Time AI-Native UI Design
Google Labs' Stitch is a strong design-time GenUI case rather than a chat-only coding demo. In its May 19, 2026 official update, Google describes Stitch as a live collaborative design system where an agent streams UI work directly onto an infinite canvas, lets users steer iterations before they finish, and exports the resulting screens toward production.
Efficient Personalization of Generative User Interfaces
Recent arXiv paper on one of the core unsolved GenUI problems: personalization. The authors collect pairwise judgments from trained designers over the same 600 generated UIs, show that design preferences diverge substantially, and propose a sample-efficient preference model that personalizes generated interfaces better than baseline evaluators and direct prompting.
MAIC-UI: Making Interactive Courseware with Generative UI
Recent arXiv paper introducing a zero-code GenUI system for educational courseware. MAIC-UI turns textbooks, PPTs, and PDFs into interactive learning pages, uses a generate-verify-optimize pipeline for pedagogical accuracy, and supports click-to-locate incremental edits instead of expensive full regeneration.
A2UI v0.9: The New Standard for Portable, Framework-Agnostic Generative UI
Official Google Developers Blog update on A2UI v0.9. The post positions A2UI as a production-oriented standard for declarative, agent-generated UI, adds the official React renderer, and explains how agents can target existing component catalogs across web and mobile without generating arbitrary frontend code.
Stop making AI write JSON - Why we built OpenUI
OpenUI's official engineering write-up on why the team moved beyond JSON trees for Generative UI. It lays out the runtime requirements for interactive agent-generated interfaces, argues that nested JSON becomes fragile for stateful UI, and introduces OpenUI Lang as a streaming-first language for model-generated UI.
Software as Content: Dynamic Applications as the Human-Agent Interaction Layer
2026 arXiv paper arguing that chat alone is a poor medium for structured, stateful human-agent work. The authors propose Software as Content, where dynamically generated agentic applications become the primary interaction layer, persist across turns, and expose actionable controls instead of forcing everything through linear text.
Introducing Tambo 1.0
Official Tambo launch post for its production-ready React generative UI toolkit. The post explains the fullstack model: developers register existing React components, the agent selects and streams props into them, and Tambo handles conversation state, orchestration, and persistent interactive component behavior.