Google Gemini adds Daily Brief and Neural Expressive redesign

At Google I/O 2026, the company unveiled a substantial set of upgrades for its Gemini app, anchored by a new Daily Brief feature and a visual refresh called Neural Expressive. The announcements, made during the opening keynote, mark a decisive shift in strategy: Google is repositioning Gemini as a proactive assistant rather than a simple reactive chatbot.

The headline feature, Daily Brief, functions as a personalised morning digest. It pulls data from a user’s inbox, calendar, and task list to create a prioritised overview of the day. More than just a summary, the tool actively suggests next steps, surfacing the most urgent items first. This feature is rolling out today to Google AI Plus, Pro, and Ultra subscribers in the United States.

This update comes as Gemini’s user base has experienced explosive growth. Google reported that the app now serves more than 900 million monthly active users across over 230 countries and 70 languages. That figure is more than double the roughly 400 million users reported at last year’s I/O, making it, by Google’s measure, the most widely available generative AI tool globally.

Alongside the functional update, Google introduced a new design language for the app. Called Neural Expressive, the redesign features fluid animations, vibrant colour palettes, updated typography, and haptic feedback. Responses are no longer dense blocks of text. Instead, key information is bolded at the top, with an option to scroll for more detail. When relevant, the app now inserts inline images, narrated videos, timelines, and interactive visualisations instead of prose.

The redesign is rolling out now on Android, iOS, and the web. It also integrates Gemini Live, the voice conversational interface, directly into the core experience. Users can switch between typing and speaking without losing their place in the conversation. For Google, this overhaul is about making AI interactions feel less like a search query and more like a conversation with an assistant that values presentation as much as content.

The most ambitious new feature is Gemini Spark, a cloud-based AI agent built on the new Gemini 3.5 Flash model. Spark is designed to handle tasks proactively across Gmail, Docs, and other connected Google services. Crucially, it continues working even after a user locks their phone or closes their laptop, since it runs entirely on Google Cloud infrastructure. No local device needs to stay active.

Spark will be available as a beta to trusted testers this week, with a wider rollout to US-based Google AI Ultra subscribers starting next week. The Ultra subscription itself has seen a significant price adjustment, dropping from $250 per month to $100 per month. This move sharpens Google’s competitive edge against OpenAI’s ChatGPT and Anthropic’s Claude. The $100 Ultra tier includes five times the usage limits of the $20 AI Pro plan, 20 terabytes of cloud storage, YouTube Premium, and beta access to Spark.

Google also announced Gemini Omni, a new AI video model that accepts images, audio, and text as inputs to generate video. Omni was spotted in the Gemini interface earlier this month, when a UI string referencing the model leaked ahead of I/O. The model is expected to roll out to Google Flow and YouTube Shorts, giving creators multimodal video tools directly within platforms they already use.

The Omni announcement fits into a broader arms race in AI-generated video. Google is competing not only with OpenAI but also with ByteDance’s Seedance and other emerging players. Early assessments suggest Omni excels at prompt adherence and in-chat editing, though its raw generation quality in the initial Flash tier may lag behind some rivals.

Taken together, these updates signal that Google is shifting Gemini from a reactive chatbot into something closer to a proactive personal operating system. Daily Brief handles the morning routine, Neural Expressive makes the interface less clinical, and Spark promises to keep working autonomously around the clock. This approach mirrors what Meta’s Mark Zuckerberg is pursuing with his own AI agent ambitions, and what OpenAI has been telegraphing with its operator-style features.

Whether 900 million users will embrace a morning briefing from their AI assistant, or whether Spark’s autonomous task execution will raise more privacy questions than it answers, remains to be seen. But with Google now embedding Gemini into everything from factory robots to mobile apps, the company is clearly betting that the future of AI is not a single chatbot but an interconnected layer that runs across every surface of daily life.