Google I/O 2026 Roundup: Gemini 3.5, AI Search, Android XR Glasses, and More

At a glance:

• Google launched multiple Gemini models including Omni, Omni Flash, and 3.5 Flash at I/O 2026
• AI Search now uses Gemini 3.5 Flash with a redesigned search box supporting multiple input types
• Android XR Audio Glasses coming this fall with private audio responses

Key AI Announcements

Google's I/O 2026 event focused heavily on AI advancements, with multiple new models and features announced. The company unveiled Gemini Omni, a new model capable of creating content from any input, with enhanced capabilities for simulating gravity and kinetic energy. This model combines Gemini intelligence with generative models like Nano Banana and Veo, supporting conversational language video editing. Users will be able to upload videos and edit any element within them.

Gemini Omni Flash is the first Omni model being released, available starting today in the Gemini app. Meanwhile, Gemini 3.5 Flash combines "frontier intelligence with action" and performs better across almost all benchmarks compared to 3.1 Pro, according to CEO Sundar Pichai. This model is described as "comparable to the best models" but faster, and is available for everyone today across Google's products and APIs. Google is also testing Gemini 3.5 Pro internally, with plans to release it next month.

The Gemini app has been redesigned with a new Neural Expressive design language rolling out today on desktop, iOS, and Android. It features fluid animations, vibrant colors, haptic feedback, and new typography. Custom regional dialects are planned for the next few months. Gemini Omni is available for paid Plus, Pro, and Ultra subscribers in the app, making it easier to create and edit videos.

AI Agents and Product Integration

Google is introducing AI agents for Gemini, including a Daily Brief agent that presents users with a customized daily digest. This feature is rolling out today for paid subscribers. The company is also enhancing the Gemini Mac App, allowing users to select images and documents in Finder and press the Function key to give Gemini voice commands. A demo showed sending an email to a dog kennel with dog information and image, pulled from Finder to generate an email using Gmail in Chrome. Voice support and Gemini Spark are coming to the Mac app this summer.

Gemini for Science, a collection of science tools, includes a Co-Scientist collaborative AI research partner. This demonstrates Google's commitment to applying AI to specialized domains beyond general-purpose applications.

Content Identification and Verification

Google is addressing the challenge of AI-generated content by implementing C2PA content credentials in both Gemini and Chrome. The company's tools can determine whether an image was captured with a camera or created with AI, and can detect if a camera-captured image was edited with AI. Users will be able to right-click on an image in Chrome and ask Gemini whether it was generated with AI, providing transparency in an era of increasingly sophisticated AI image generation.

Antigravity 2.0 Coding Tool

Google is launching Antigravity 2.0, a new agent-first app for desktop that uses Gemini 3.5 Flash. Antigravity is Google's coding tool, positioned as a competitor to Microsoft's Copilot, GitHub's Codex, and Anthropic's Claude Code. The new version is 12x faster than previous iterations, optimizing token use. Antigravity 2.0 is available globally for everyone, suggesting Google is making this powerful coding assistance tool broadly accessible.

Gemini Spark Personal Agent

Gemini Spark is a personal AI agent designed to help users navigate their digital life. It runs on virtual machines through Google Cloud, operating 24/7 without requiring a laptop to be open. Accessible through the Gemini app, it will also support email and messaging options. The agent uses Gemini 3.5 Flash and Antigravity to work on long-running tasks in the background, integrating with Google tools now and debuting MCP support for third-party apps in coming weeks.

Gemini Spark can handle multi-step ongoing tasks, planning subtasks and executing them sequentially. It will be available for Google AI Ultra subscribers in the U.S. next week, with Chrome integration coming later this summer.

Google Search Transformation

Google Search is undergoing a significant transformation with AI Search now using Gemini 3.5 Flash. The company's messaging is clear: "Google Search is AI Search." A new search box has been reimagined with AI capabilities, changing based on usage patterns and going "beyond autocomplete" to better assist with questions. The search box now supports images, files, videos, and Chrome tabs as input in addition to text. Pichai described this as the biggest upgrade to the search box in over 25 years, with the feature rolling out today.

Google is also combining AI Overviews and AI Mode into a single interface. Search is gaining support for creating and managing multiple AI agents that can keep users updated on real-time information like stock market changes and send alerts. The system can monitor webpage changes, providing alerts for events like sneaker drops. Information Agents are coming to search in the summer, with Google planning to expand the agent ecosystem.

Agentic Search Capabilities

Google is introducing agentic coding capabilities in Search, allowing it to build custom responses on the fly with dynamic layouts and interactive widgets. This functionality uses Antigravity and Gemini 3.5 Flash. Search can create tools, trackers, widgets, and dashboards based on user queries. The Generative UI in Search is rolling out this summer for everyone at no charge, while Antigravity in Search for building custom experiences is coming in the summer for subscribers first.

Universal Shopping Cart

Google is introducing a Universal Cart coming this summer to Search and the Gemini app. This intelligent shopping cart works across merchants and services, allowing users to add items when reading Gmail, watching YouTube, or browsing the web, then check out on Google or third-party retailer sites. Google has developed a Universal Commerce Protocol and an Agents Payment Protocol for agentic shopping. The payments feature allows AI agents to make payments on behalf of users using parameters like brand and price preferences. This functionality will be integrated with Gemini Spark later this year.

YouTube and Document Enhancements

YouTube is getting an "Ask YouTube" feature similar to the Ask Maps AI functionality. It uses Gemini to allow users to ask questions with context and follow-up capabilities. The feature is currently in testing and will roll out broadly in the United States this summer.

Google Docs is enhancing its Live feature with AI integration. Users can speak or write parameters of what they need, and Gemini can create documents. Pichai described this as allowing users to "brain dump" and then let Gemini "do the rest." The feature supports text-based commands for creating and editing content.

Hardware Innovations

Google is entering the wearable space with Android XR Audio Glasses coming this fall. These glasses provide all-day access to Gemini with responses privately spoken into the wearer's ear. They can be used for taking photos, listening to music, making calls, and accessing apps. Google collaborated with Gentle Monster and Warby Parker on the design, and Samsung on hardware development. The glasses will pair with both Android and iOS devices, expanding Google's ecosystem beyond its own platform.

Additional Product Updates

Google has announced a new $100 Ultra plan, reducing the price of its top-tier Ultra plan from $250 to $200. Android users will get Android Halo, which lets them keep an eye on AI agents with activity displayed at the top of their devices. This feature is coming to Android later this year.

Google Pics is a new image generation and editing tool in Google Workspace, similar to Canva, allowing users to create posters, flyers, infographics, and more. Content is watermarked with SynthID for identification. Pics is rolling out this summer.

Stitch is Google's Figma-like tool for building apps and websites. This year, Google is adding real-time collaborative design with Stitch Agent, exports to Antigravity, and direct publishing to Netlify.

Google Flow, Google's "AI creative studio" for creatives, is getting Gemini Omni, AI agents for executing multiple actions simultaneously, and custom tools with Flow Tools. These new Google Flow features are available today.

Competitive Context

Many of Google's new features are rolling out today, with others planned for later this year. The announcements set the stage for Apple's WWDC 2026 keynote on June 8, where the tech world will see if Apple debuts equivalent features. Google's focus on AI integration across its product line demonstrates its commitment to maintaining its competitive edge in the rapidly evolving AI landscape.