Top 13 Highlights from Google I/O 2026
Google’s I/O 2026 keynote was packed with exciting AI announcements, including the introduction of the Gemini 3.5 AI models, fresh features for Search and Gmail, and updates on the Project Aura smart glasses. If you missed the livestream or our live updates, here’s a recap of the key highlights.
**Gemini 3.5**
Google unveiled its latest AI models, kicking off with Gemini 3.5 Flash, with Gemini 3.5 Pro set to launch next month. Starting today, Gemini 3.5 Flash becomes the default model for the Gemini app and AI Mode in Search. This new model boasts significant speed improvements, enhanced capabilities for handling agentic tasks, better coding features, and the ability to create “richer, more interactive web UIs and graphics.” Additionally, it comes with improved safety measures to reduce the chances of generating harmful content and misclassifying safe queries.
Alongside Gemini 3.5 Flash, Google is rolling out a redesign of the Gemini app, featuring a “neural expressive” design with new animations, vibrant colors, a fresh font, and haptic feedback. This update will be available starting May 19th on the web and in the Gemini app for Android and iOS.
**Gemini Omni**
In addition to Gemini 3.5, Google introduced a new family of AI models called Gemini Omni. The first model, Omni Flash, is now available in the Gemini app, Google Flow, and YouTube Shorts. Omni Flash can generate video clips from various inputs, including text, photos, video, and audio, unlike the previous Veo model, which was limited to text-to-video. Google envisions that Omni will eventually be able to “create anything from any input.”
**Gemini Spark**
Gemini Spark is Google’s answer to OpenClaw, functioning as an always-on AI assistant that can draft emails, create study guides, and monitor for hidden credit card fees. Powered by Gemini 3.5 Flash, Spark operates continuously in the background using virtual machines on Google Cloud and integrates with Google Workspace apps like Docs, Gmail, Sheets, and Slides, as well as third-party applications like Canva and Instacart. Future updates will allow Spark to access local files through the Gemini app on macOS.
**Build Android Apps with Google AI Studio**
Google is enabling users to vibe-code entire native Android apps, which can eventually be published directly to the Play Store from AI Studio. This feature includes an embedded Android emulator for previewing and editing apps. Users can connect their phones to install apps directly for testing, and apps can be exported to Android Studio, GitHub, or saved as ZIP files. Google also plans to allow users to publish vibe-coded apps privately for friends and family, with Firebase integration coming soon.
**Project Aura Smart Glasses**
Google showcased an updated version of its Project Aura smart glasses, developed in collaboration with Xreal. Originally revealed last year, the updated model features a redesigned external compute puck with a fingerprint sensor and a lanyard for easy wear. Attendees experienced new features in Google’s XR platform, including widgets for display glasses and improved Gemini performance with integrations for Google Calendar and Google Keep.
**New Android XR Glasses**
This fall, two new pairs of Android XR smart glasses will launch, one from Warby Parker and another from Gentle Monster. These audio-only glasses will support features like live translation, navigation assistance with Gemini, and notification summaries.
**Universal Cart**
Google is introducing a “Universal Cart” that allows users to add products from YouTube, Search, Gemini, and Gmail. This intelligent shopping cart works across various merchants, enabling users to check out from multiple stores simultaneously. It can also identify potential issues with products in the cart, such as incompatible parts for a gaming PC, and interpret loyalty information from Google Wallet to help users save or earn points. The Universal Cart will launch in Search and Gemini this summer, with YouTube and Gmail to follow.
**Gmail Live**
Gmail is enhancing its search capabilities with a version of the Gemini Live experience. Users can now search their inboxes using voice commands, allowing for quicker access to specific information, like hotel confirmation codes. Similar voice-driven features will also be available in Google Docs and Keep, pulling data from Google Drive and Gmail.
**Pics App for Image Editing**
Google Workspace is introducing a new app called Pics, designed to simplify the process of applying iterative AI-generated updates to images. Powered by Nano Banana 2 and Gemini, users can make changes by clicking on parts of an image and leaving comments, rather than crafting full prompts. Google plans to integrate Pics’ features into other Workspace apps for streamlined editing.
**Search Enhancements**
Google is revamping the search box to allow for longer queries and AI-generated suggestions. Users can now search using text, images, files, videos, and even Chrome tabs. The search feature will also introduce “information agents” that provide summarized updates on specific topics, launching this summer for AI Pro and Ultra subscribers. Additionally, a new “generative UI” feature will create visuals in Search, such as simulations and interactive graphs, and mini apps for frequently searched tasks.
**AI Ultra Plan Pricing**
Google is reducing the price of its AI Ultra subscription, which previously cost $249.99 per month. The new pricing starts at $100 per month, with a $200 option that includes access to Google’s Project Genie.
**AI Detection Tools**
To help users identify AI-generated or altered images, Google is expanding its AI detection tools to Chrome and Search. Users can upload or select images to reveal details about their creation using Google’s SynthID watermarking technology and C2PA Content Credentials. Additional functionality for Chrome will allow users to check the provenance of questionable images on websites.
**Google Beam AI Agents**
After rebranding Project Starline to Google Beam, the company is now testing lifelike AI agents that enhance interaction during calls. An early demo showcased Sophie, a Beam video agent capable of responding to questions, reading documents, and providing recommendations. Group calls using Beam, compatible with Google Meet and Zoom, are also in development.
