Google integrates AI into all key services: Highlights from I/O 2025

At the Google I/O 2025 conference, the company unveiled a series of AI-focused innovations, including updates to Gemini, new search features, and content creation tools.

AI Mode in Google Search

Google introduces AI Mode in its search engine, allowing users to interact with the Gemini chatbot for deeper answers, including deep search capabilities, chart generation for financial and sports queries, and the ability to make purchases directly through AI Mode.

Gemini 2.5 Pro Updates

The Gemini 2.5 Pro model has received a Deep Think mode designed to tackle complex tasks such as mathematical calculations and programming. Additionally, Gemini integrates into the Chrome browser, enabling information summarization on web pages and task execution directly within the browser.

Smart replies in Gmail

Gmail is receiving enhanced smart replies that utilize AI to consider context from incoming emails and Google Drive, providing more personalized and accurate responses. This feature will be available in alpha version in July 2025.

Xreal + Google = Project Aura

Xreal showcased a prototype of new smart glasses, Project Aura, developed in partnership with Google.

• Externally resemble stylish sunglasses.
• Operate on the Android XR platform.
• Equipped with cameras, microphones, and a wide field of view.
• Integrate Gemini and mixed reality features.

While still a prototype, they already appear as potential competitors to Apple Vision Pro.

AI for shopping

Google is testing a new AI-powered clothing try-on feature – upload a full-length photo of yourself and see how selected outfits would look on you. The AI considers body features, fabric, and fit. An "agent checkout" feature is also forthcoming, where AI will independently make purchases on your behalf.

New content creation tools

• Imagen 4: An improved image generator with better quality and support for various formats.
• Veo 3: A model for generating video and audio with new tools allowing simultaneous generation of video and sound.
• Flow: A new app that uses Veo, Imagen, and Gemini to create eight-second AI-generated video clips based on text prompts and/or images. It also comes with scene creation tools to connect clips and create longer AI-based videos.

Other important announcements

• Google Beam: Rebranding of Project Starline, a new device for 3D video communication, which will be integrated into HP devices.

• AI Ultra: A premium subscription to Google's AI services for $250 per month, offering access to the company's most advanced AI models and higher usage limits in applications like Gemini, NotebookLM, Flow, and others.
• Stitch: A new AI-based tool that can create interfaces using selected themes and descriptions. You can also add wireframes, sketches, and screenshots of other interface designs to guide Stitch's results. The experiment is currently available in Google Labs.