Google I/O 2026 saw the launch of the new Ask YouTube feature, bringing natural language search to the video streaming platform.
Here's Everything That Was Announced During the Google IO 2026
Photo Credit: Google
Google I/O 2026, the latest edition of the Mountain View-based tech giant's developer conference, was hosted by the company on Friday. During the keynote, Google's head Sundar Pichai announced various new AI features and tools for Google products. Moreover, the company executive revealed the number of tokens the tech giant is now processing per month to provide AI-generated results for its users, which is up seven times from the same month last year. On top of this, the tech giant has introduced more Gemini-powered tools across its products, including YouTube, Gmail, and Search. The company has also released its latest Gemini 3.5 Flash model.
During the Google I/O 2026 developer conference, Pichai revealed that Google now processes more than 3.2 quadrillion tokens per month, which is claimed to be a seven times increase from the same period last year. For reference, the tech giant was processing 480 trillion tokens per month by May 2025, while processing 9.7 trillion tokens per month in the same period in 2024.
Moreover, the company has also unveiled its Gemini 3.5 series models, starting with the Gemini 3.5 Flash, which is currently available to all users globally in the Gemini app, AI Mode in Search. Meanwhile, developers can use the model in the Antigravity platform and in the Gemini application programming interface (API) via the AI Studio and Android Studio. Meanwhile, the company plans to release the Gemini 3.5 Pro model next month.
The new Gemini 3.5 Flash model is claimed to be four times faster than comparable frontier models. Meanwhile, it is positioned to cost significantly less in AI costs. Google claimed that Gemini 3.5 Flash was able to generate an operating system in 12 hours.
Coming to Google's new video generation model, Gemini Omni can combine multimodal inputs into a single prompt to generate a video. The tech giant highlighted that users can input a reference video, image, text, and audio input within a prompt, explain the shot and the camera angle, and Gemini Omni will generate a video for them. It is now available to Google AI Plus, Pro, and Ultra users globally via the Gemini app and Google Flow.
Additionally, Google has also launched the new Gemini Spark tool, which is positioned as the company's new “personal AI agent”. Powered by Google's Gemini 3.5, the new tool is reportedly capable of taking action on a user's behalf and performing “long-running” tasks in the background. Users do not have to keep their devices open to keep the agent running, as it runs on dedicated virtual machines on Google Cloud. It will be integrated into other Google tools, too.
Google's Gmail, Google Docs, and Google Keep are getting upgraded with the new AI tools, allowing users to initiate natural language queries. Users can ask Gemini in the Google Workspace apps to brainstorm on a topic of their choice or organise their thoughts using text and voice commands. Additionally, the company is also integrating Gemini Spark into these platforms. The features will be rolled out to users in the summer.
Moreover, Google Pics has been introduced as the tech giant's image creation and editing tool, which is powered by Google's Nano Banana model. It is capable of object segmentation, text editing, and translation. Google Pics will be available in Google Workspace platforms.
Lastly, the company is explaining the AI Inbox functionality in Gmail to Google AI Plus and Pro subscribers with new features like personalised draft replies and instant file access, along with the ability to mark individual tasks as done, dismissing unhelpful suggestions, and marking all emails in a given topic as read with a single click.
Google has introduced a new way for users to search on its video streaming platform. Dubbed Ask YouTube, the new functionality lets users ask more complex questions to the platform. For example, users can search for an exact part in a video, even if they can not recall the exact name of the video. Ask YouTube is also capable of compiling the “most relevant videos”, retrieving details from across YouTube's catalogue, including YouTube videos and YouTube Shorts, to generate responses.
Ask YouTube is currently being rolled out to YouTube Premium subscribers in the US who are aged 18 and above. The company plans to release the feature to other regions soon.
Another major highlight from the event was the Google Search upgrade. The tech giant has introduced a new AI Search box as an extension of its AI Mode in Search. It offers a dedicated space to users for doing conversational searches in a single place. Now, Google Search will also present AI-generated suggestions for search questions to users, while also supporting multimodal queries. Hence, users can use images, text, and audio inputs to search.
Moreover, it will be capable of managing multiple AI agents simultaneously. However, initially, it will only support information agents, which can operate in the background, allowing users to stay updated with various topics.
Google Search has also been upgraded with agentic booking capabilities, along with agentic coding functionality. For example, Google Search can now find a private karaoke room for six on a Friday night that serves food late and present the pricing and availability with direct links to the booking platforms. The agentic capabilities are scheduled to be released to everyone in the US later in the summer.
The Mountain View-based tech giant is also bringing agentic commerce capabilities to different platforms with the help of the Universal Commerce Protocol (UCP). Dubbed Universal Cart, the tool will allow users to shop while using Google Search, within the Gemini app, while watching a video on YouTube, or while reading emails in the Gmail app. The company claims that the tool works across merchants and services.
Google has announced that it will roll out the new Android Halo indicator with the Android 17 update later this year. The new tool will display to users the tasks that the AI agent is performing at the time. The indicator will appear in the top-left corner of the screen next to the time.
Lastly, the company showcased the new smart glasses, which will be based on its Android XR platform. Currently, Samsung's AR/VR headset is the only device running the Android XR OS. While the company last year confirmed that it would launch new smart glasses, now it has confirmed the availability of two categories of intelligent wearables, including an audio glasses model and another model with a display.
Get your daily dose of tech news, reviews, and insights, in under 80 characters on Gadgets 360 Turbo. Connect with fellow tech lovers on our Forum. Follow us on X, Facebook, WhatsApp, Threads and Google News for instant updates. Catch all the action on our YouTube channel.