13 Things Google launched at I/O 2025

Here are 13 key updates from Google I/O 2025 with a focus on practical applications like filmmaking tools, image generation, and enhanced search capabilities. Gemini has been given big updates and more integration into Google’s apps.

Google says across Africa, more and more people are finding themselves interacting with artificial intelligence every day, perhaps without even realising it.

“From the personalised recommendations that pop up on their phones to the smart assistants helping them manage their schedules, AI is quietly becoming an indispensable part of modern life. This widespread adoption underscores a global trend: AI is evolving from a complex theoretical concept into practical, helpful tools that solve real-world problems and enhance our capabilities.”

Here are 13 AI-powered tools from I/O 2025:

1. Gemini 2.5 AI Model

Gemini 2.5 Flash is now the default model with fast response times, offering speed and efficiency for daily interactions. For enhanced reasoning and tackling complex challenges, an experimental mode called Deep Think is being explored in Gemini 2.5 Pro

2. Imagen 4 Image Generation

Imagen 4, an image generation model is now accessible in the Gemini app. Its richer, more nuanced images and fine-grain details bring visual ideas to life with clarity. Imagen 4 has improved text and typography, making creative font, spacing, and layout choices. A faster variant, up to 10x quicker than the previous model, is coming soon, enabling faster idea iteration and boosting creativity.

3. Veo 3 Video Generation

Google’s video generation model, Veo 3 was launched alongside Imagen 4. Available in the Gemini app for US subscribers, Veo 3 offers improved visual quality, realistic animations, and intuitive controls. It also includes native audio generation, allowing users to add sound effects and background noise, with dialogue creation coming soon.

4. Flow Filmmaking

Flow, an AI filmmaking tool built for creatives, helps users create cinematic clips, scenes, and stories with consistency. It uses Gemini models for intuitive prompting and allows users to create story elements like cast, locations, objects, and styles using natural language. Users can also integrate assets and reference images for consistency. Flow is available for Google AI Pro and Ultra subscribers in the US only.

5. AI Mode in Search

AI Mode in Search launches in the US first. It features advanced reasoning and multimodality, enabling deeper exploration with follow-up questions and helpful links. Gemini 2.5 powers AI Mode and AI Overviews in the U.S. as of last week. Many AI Mode features will integrate into the core Search experience over time.

6. Gemini in Chrome

Gemini in Chrome is rolling out on desktop to Google AI Pro and Google AI Ultra subscribers in the US who use English as their primary Chrome language on Windows and macOS. It lets you easily ask Gemini to clarify complex information or summarise it on any webpage. In the future, it’ll work across multiple tabs and navigate websites for you.

7. Search Live

Search Live, a new feature in Search, enables users in the US to engage in real-time conversations with Search using their cameras. It functions as a learning partner, explaining concepts or offering suggestions based on visual input, with links to other resources. Search Live is expected to launch in AI Mode in Labs this summer.

8. Tasks with AI Mode

Google integrates Project Mariner’s capabilities into AI Mode in the US to save users time by automating tasks like event ticket purchases, restaurant reservations, and local appointments. AI Mode scans sites, analyses options, handles tedious form-filling, and presents relevant choices while completing the purchase on the user’s preferred site. These agentic capabilities arrive in AI Mode Labs this summer.

9. Deep Research AI Mode

Deep Search, an advanced “query fan-out” technique, will arrive in AI Mode in Labs this summer. It can initiate hundreds of searches, reason across information, and create expert-level, fully-cited reports in minutes, potentially saving users hours of research and helping them grasp complex topics quickly.

10. Personalised Search Results

AI Mode in Labs will offer personalised suggestions based on past searches, enhancing the search experience. Users can optionally connect other Google apps, starting with Gmail, for tailored responses, like restaurant suggestions based on booking history. This feature is user-controlled, allowing connection or disconnection anytime. AI Mode will also analyse complex datasets and create custom charts and graphs for data visualisation in sports and finance queries, arriving this summer for clearer information understanding.

11. Smart Shopping

The new AI Mode shopping experience combines Gemini model capabilities with Google’s Shopping Graph to help users browse, consider options, and narrow down products efficiently. It includes a “try-on” feature, allowing users to virtually try on billions of apparel listings by uploading their image, starting with Labs users in the US today. Users can also ask an agentic checkout feature to make purchases with Google Pay when the price is right, with their guidance. This experience is expected to launch in AI Mode in the U.S. in the coming months.

12. Gemini App Enhanced

To make Gemini a more helpful assistant, Google integrates capabilities from Project Astra, like video understanding and improved memory, enhancing interactions and productivity. Gemini Live will soon integrate with Google services for seamless daily assistance. iOS users can now use their camera or share their screen in Gemini Live, along with Android users. More Project Astra live capabilities are coming soon.

13. Agent Mode for Complex Tasks

Google AI Ultra subscribers will soon get an experimental Agent Mode in the Gemini app. This lets users delegate complex planning and tasks, combining features like live web browsing, in-depth research, and Google app integrations to manage multi-step tasks efficiently, freeing up time and mental energy.