Google used its Android Show: I/O Edition 2026 event to unveil a sweeping suite of AI-powered features under the Gemini Intelligence banner, signalling its most ambitious push yet to embed generative AI throughout the Android experience. The features will initially roll out to Samsung Galaxy and Google Pixel devices this summer before reaching other Android phones later in the year.
At the centre of the announcement is Rambler, a new AI voice dictation feature built directly into Gboard. Powered by Gemini-based multilingual models, Rambler removes filler words, handles mid-sentence corrections, and supports code switching — allowing users to move fluidly between languages within a single sentence. Ben Greenwood, director of Android Core Experiences, described the capability as effectively reinventing the keyboard. Google said voice data is not stored and is used solely for transcription, with processing split between on-device and cloud infrastructure.
Google also introduced Create My Widget, which lets users vibe-code personalised Android home screen dashboards using plain natural language. Gemini can draw on web data alongside personal information from Gmail and Calendar to build dynamic, resizable widgets tailored to individual needs.
Beyond those headline features, Gemini is gaining broader agentic powers — completing multistep tasks across apps, auto-browsing the web, filling out forms through opt-in Personal Intelligence, and summarising web pages via Gemini in Chrome, coming to Android in late June. Greenwood described the vision as a personal assistant that continuously surfaces relevant answers from both the world and a user’s own data.
