Google Disco: Build Apps Instantly
Turn chaotic browser tabs into intelligent AI apps that work for you instantly.
26 mar 2026 (Aggiornato il 26 mar 2026) - Scritto da Christian Tico
Source: Google.
Perché le tue idee migliori falliscono (E come evitarlo)
Smetti di lanciare campagne marketing o video basandoti solo sull'istinto. Testa i tuoi concetti sul Pannello di Valutazione delle Idee IA per scoprinre subito punti di forza e debolezze.
Google's AI Browser Revolution: Generating Websites in Real Time
Google is pushing the boundaries of web browsing with experimental AI tools that transform open tabs into dynamic websites and interactive apps on the fly. These innovations promise to redefine how we interact with information, turning chaotic browsing sessions into seamless, real-time generated experiences.
Disco: Turning Browser Tabs into Custom Websites
One standout Google Labs experiment, Disco, analyzes your open tabs to understand your intent and generates tailored mini-apps instantly. Instead of sifting through dozens of tabs, it creates visual workspaces called GenTabs that consolidate everything into one intuitive screen.
- Competitor sites become comparison matrices for quick side-by-side analysis.
- Travel tabs evolve into personalized itinerary planners.
- Recipe collections turn into meal planners complete with shopping lists.
Every element links back to original sources, ensuring accuracy while delivering a cohesive, app-like interface generated in real time.
Project Mariner: AI Agents That Build and Navigate
Complementing Disco, Project Mariner introduces an AI agent that browses the web visually, mimicking human actions with a ghost cursor you can observe in real time. It identifies buttons, search bars, and forms across sites, opens multiple tabs, and completes complex tasks autonomously.
Users can intervene by clicking "Take Over" at any point, blending AI efficiency with human oversight. This capability extends to generating structured outputs from web data, effectively building simple websites or dashboards on demand.
Gemini Integration and AI Studio Enhancements
Gemini now integrates directly into browser side panels, enabling multitasking without tab switching. It summarizes pages, compares options across tabs, and even transforms images using tools like Nano Banana.
Google AI Studio takes this further by letting users describe desired apps in plain language. The AI handles code generation, API connections, and instant deployment to public URLs. Edit elements directly by typing changes, making real-time website creation accessible to non-coders.
Broader Context: AI Browsers in 2026
Google's efforts come amid a surge in AI browsers, with competitors like Perplexity's Comet introducing smart tab management and predictive loading. Chrome's Gemini extension, currently U.S.-only, hints at agentic features in development that could solidify its market lead.
These tools emphasize local-first execution and seamless AI assistance, shifting browsers from passive viewers to proactive creators that generate websites tailored to user needs.
Conclusion
Google's demos showcase a future where AI browsers not only organize information but generate fully functional websites in real time from everyday browsing. As these experiments mature, they could make web creation effortless for everyone.
This evolution signals a transformative shift, empowering users to build without coding while maintaining control over their digital workflows.
Google’s AI browser experiments make it clear that AI is not just for “finding pages”, but for instantly building the right interface to study what you already have open: comparisons, outlines, planners, dashboards. Once you start using it this way, asking it to turn tab chaos into matrices, summaries, or mini‑apps, the browser stops being a place where you get lost and becomes a genuine learning environment, helping you understand faster and with far less friction.
How does Project Mariner function as an AI agent during a browsing session?