Karachi, Pakistan, May 21, 2025: At Google I/O, Google unveiled its latest advancements in Artificial Intelligence, showcasing how cutting-edge research is being brought to reality to build the most helpful AI. The focus is on creating AI that is more intelligent, agentic, and personalized for everyone, everywhere. This marks a new phase in the AI platform shift, where decades of research are becoming reality for people, businesses, and communities globally. The announcements preview Google’s vision and several big updates. At this Google I/O, key announcements showcase how AI is transforming various Google products and experiences. “More intelligence is available, for everyone, everywhere. And the world is responding, adopting AI faster than ever before… What all this progress means is that we’re in a new phase of the AI platform shift. Where decades of research are now becoming reality for people, businesses and communities all over the world.” — Sundar Pichai, CEO, Google and Alphabet.
These key announcements from the event showcase how AI is revolutionizing various Google products and experiences.
- Gemini 2.5 Upgrades
Gemini 2.5 introduces a new era of advanced performance with features like Deep Think, an enhanced reasoning mode designed for complex tasks like coding and math. Developers now benefit from thought summaries to better understand how the model processes tasks, as well as extended thinking budgets in Gemini 2.5 Pro — offering more control and customization. The model now also supports native audio generation and computer use, pushing the boundaries of what AI can do.
- Smarter Google Search
Google Search has entered its next phase with AI Mode, a new interface powered by Gemini that helps users explore topics more deeply and interactively. AI Overviews — already saving users over a billion clicks per month — are now live in 200+ countries and over 40 languages. With new features like video understanding and screen sharing rolling out, Google is transforming Search from a source of information into an engine of intelligence.
🔗 AI in Search
🔗 AI Overview Expansion - Generative AI Media Tools
To empower the next generation of creators, Google announced its most advanced generative media models: Imagen 4 for high-fidelity images, Veo 3 for cinematic video, and Lyria 2 for AI-generated music. These tools are now accessible through Flow — Google’s new AI-powered filmmaking platform designed to help creators storyboard, direct, and produce short clips with natural language prompts and media references.
🔗 Explore the tools
🔗 Meet Flow - Gemini App Updates
The Gemini app is evolving into a more intuitive assistant. Now integrated with camera input, screen sharing, and world modeling from Project Astra, Gemini can better understand your surroundings and anticipate what you need next. These upgrades allow Gemini to assist across tasks — whether you’re solving problems, creating content, or navigating your digital life — in real time, across devices.
🔗 See what’s new - Workspace with Gemini
Gemini is now deeply embedded into Workspace, bringing personalized smart replies to Gmail, real-time speech translation in Google Meet, and research-backed writing in Docs. The new Vids app lets users create short videos using slides, voice input, and AI editing. With these tools, Workspace becomes a productivity suite that helps users communicate faster, collaborate smarter, and stay focused on high-impact work.
🔗 Learn more
These updates reflect Google’s commitment to building a universal AI assistant that’s fast, helpful, and widely accessible.
For queries, reach out to your local PR contact or press@google.com.