Google unveiled a wide range of artificial intelligence products at its I/O 2026 developer conference, pushing Gemini deeper into Search, Workspace, YouTube, Chrome, Android and developer tools as the company moves to turn AI from a chatbot into an agent that can carry out tasks across daily digital life.
The announcements marked Google’s clearest effort yet to position Gemini as the central AI layer across its consumer products, cloud infrastructure and software platforms, with new models, personal agents, voice tools and content-verification features all introduced under what the company called its “agentic Gemini era.”
Google points to surging AI demand
Google said AI usage has accelerated sharply across its products, developers and enterprise customers, pointing to token processing as one measure of adoption. The company said it now processes more than 3.2 quadrillion tokens a month across its services, compared with about 480 trillion a year earlier and 9.7 trillion two years ago.
The company also said more than 8.5 million developers now build with its models each month, while its model APIs process roughly 19 billion tokens per minute. Google framed those figures as evidence that AI demand is moving beyond experimentation and becoming part of everyday consumer and business workflows.
Gemini has also helped lift engagement across major Google products. AI Overviews in Search now has more than 2.5 billion monthly active users, while AI Mode has surpassed 1 billion monthly active users. The Gemini app has grown to more than 900 million monthly active users, more than doubling from last year’s I/O.
Gemini Omni and 3.5 Flash lead model push
Google introduced Gemini Omni, a new model family designed to generate outputs across different formats from any input, beginning with video. The first version, Gemini Omni Flash, is available in the Gemini app, Google Flow and YouTube Shorts, with developer and enterprise API access expected in the coming weeks.
The company said the model combines Gemini’s intelligence with Google’s generative media systems, presenting it as a major step in world understanding as AI moves beyond predicting text and toward systems that can simulate and generate richer forms of media.
Google also launched Gemini 3.5 Flash, its latest model focused on speed, coding and agentic workflows. The company described the model as combining “frontier intelligence with action,” and said it is significantly faster than other frontier models while remaining less expensive for high-volume users.
Gemini 3.5 Flash is available across Google products and APIs, while Gemini 3.5 Pro is being used internally and is expected to arrive next month.
Gemini Spark brings agents to consumers
One of the event’s biggest consumer announcements was Gemini Spark, a personal AI agent inside the Gemini app that Google says can operate around the clock on a user’s behalf.
Spark runs on dedicated virtual machines on Google Cloud and is powered by Gemini 3.5 and Google’s Antigravity agent framework. Google said the tool is designed to handle long-horizon tasks in the background while integrating with Google services and, later, third-party tools.
The company said Spark will begin rolling out to trusted testers this week, with a beta coming next week for Google AI Ultra subscribers in the United States. Google also said Spark will later operate directly inside Chrome, effectively turning the browser into an agentic workspace.
Search, YouTube and Workspace get AI upgrades
Google also outlined major AI changes to Search, including information agents that can work in the background to find relevant updates and help users take action. The company said Search will also generate custom layouts, interactive visuals and, eventually, persistent dashboards for longer-running tasks.
In YouTube, Google is testing Ask YouTube, a feature that helps users find answers inside videos and jump directly to the most relevant sections. The tool is expected to roll out more broadly in the United States this summer.
Workspace is getting a voice-driven feature called Docs Live, which lets users verbally lay out rough ideas and have Gemini turn them into documents. Google said similar voice capabilities will also come to Gmail and Keep.
Google expands AI transparency tools
As Google expands generative AI across more products, it also announced new transparency tools. The company said SynthID has now watermarked more than 100 billion images and videos, along with 60,000 years of audio assets.
Google said it will expand SynthID and Content Credentials verification to Search and Chrome, while OpenAI, Kakao and ElevenLabs are adopting SynthID. The move suggests Google wants its watermarking technology to become part of a broader industry standard as AI-generated media becomes more difficult to identify.
The announcements show that Google is trying to position Gemini as the connective layer across its ecosystem, from search and productivity to creative tools and consumer agents. For users, the shift could make Google products more proactive and personalized; for Google, it pushes the company further into the AI race as competition intensifies.
