Google I/O 2025: All the AI & Android XR Announcements in 15 Minutes

Gemini AI Advancements

• 🤖 Gemini is evolving into a comprehensive AI season, with capabilities like completing the Pokemon game and powering new communication platforms.
• 💡 Google Beam introduces an AI-first video communication platform that transforms 2D streams into a realistic 3D experience.
• 🗣️ Real-time speech translation is now available in Google Meet for subscribers, with more languages and enterprise features rolling out.
• 🧠 Personal context allows Gemini models to use relevant Google app data privately and transparently, enhancing features like personalized smart replies.
• 🚀 Gemini 2.5 Pro and Gemini 2.5 Flash are updated with improved reasoning, coding, and long-context capabilities, becoming generally available soon.
• 🔊 New text-to-speech previews offer first-of-its-kind multi-speaker support for more expressive conversations in over 24 languages.
• 💻 Gemini 2.5 Pro demonstrates advanced coding capabilities, assisting with web app development and integrated into IDEs and Google products.
• 💡 Deep Think mode pushes model performance to its limits for groundbreaking results, available to trusted testers via the Gemini API.
• 🌍 Gemini aims to become a world model, capable of planning and simulating aspects of the world to imagine new experiences.

AI-Powered Search and Shopping

• 🔍 An all-new AI mode is rolling out to Google Search in the US, providing comprehensive, cited reports and personalized suggestions.
• 🛍️ AI mode is enhancing Google Shopping with virtual try-on features using custom image generation models for fashion.
• 📸 Search Live integrates Project Astra's live capabilities into AI mode, allowing search to see and provide information in real-time via the camera.
• 📄 Deep research now allows users to upload their own files to guide research agents and will soon support research across Google Drive and Gmail.

Creative AI Tools and Media Generation

• 🎨 Canvas transforms reports into dynamic web pages, infographics, quizzes, or custom podcasts with a single tap.
• 🖼️ Imagen 4 is the latest image generation model, available in the Gemini app, capable of creating detailed posters with creative choices.
• 🎬 Veo 3 is a new state-of-the-art model with native audio generation, creating high-fidelity video with realistic character speech.
• 🎵 Lyria 2 generates high-fidelity music and professional-grade audio with expressive vocals and rich musical arrangements.
• 💧 Synth ID watermarking technology is being updated with a detector to identify AI-generated images, audio, text, or video.
• 🎥 Flows AI Video Creator is a new tool for filmmakers that ensures character and scene consistency and allows precise camera control.

Android XR and Emerging Form Factors

• 👓 Android XR is being built to offer diverse experiences across different devices, optimized for Snapdragon with Qualcomm.
• 📱 Samsung's Project Muhan is the first Android XR device, designed as lightweight glasses for all-day wear with integrated cameras, microphones, and speakers.
• 🌐 These glasses work with phones, providing hands-free access to apps and Gemini's capabilities, including real-time communication and information retrieval.
• 🤝 Gentle Monster and Warby Parker are the first eyewear partners to build glasses with Android XR.