Google I/O Conference Goes All-in on AI: New Gemini Models, the Biggest Search Box Redesign in 25 Years, Return of AI Glasses, and Debut of Spark Agents
TechFlow Selected TechFlow Selected
Google I/O Conference Goes All-in on AI: New Gemini Models, the Biggest Search Box Redesign in 25 Years, Return of AI Glasses, and Debut of Spark Agents
Gemini Omni is positioned as Google’s most advanced multimodal AI system; Gemini 3.5 Flash emphasizes speed and low cost and will serve as the default model for the Gemini app and Search AI Mode, while Gemini 3.5 Pro will officially launch in June.
Navigating Web3 tides with focused insightsOn Tuesday, May 19, Eastern Time, Alphabet’s Google unveiled a wave of AI updates at its annual developer conference, Google I/O 2026—ranging from a complete redesign of Google Search to the launch of the Gemini Omni and Gemini 3.5 Flash models and the Gemini Spark agent, all the way to the first public unveiling of its next-generation AI glasses. Google is striving to embed AI across the entire internet gateway ecosystem—search, Android, productivity, video, and next-gen hardware.
Sundar Pichai, CEO of Alphabet and Google, revealed in his keynote address that the Gemini app has doubled its monthly active users (MAUs) year-over-year, now exceeding 900 million; “AI Overviews” in Google Search currently boast over 2.5 billion MAUs; and users of “AI Mode” have surpassed 1 billion.
Pichai emphasized that AI is no longer just about chatbots—it is “the next computing platform.” He stated at the conference: “AI is lighting up every part of our company.”
From the newly launched Gemini Omni and Gemini 3.5 Flash models, to the task-sustaining Gemini Spark agent, and even the restructured AI subscription framework, Google is transforming Gemini from a chatbot into an “AI operating system” spanning search, productivity, development, video, and mobile ecosystems.
This year’s conference is widely seen as Google’s comprehensive counteroffensive against competitors including OpenAI and Microsoft. Over the past year, as ChatGPT reshaped user search and content-consumption habits, Google’s core search business has faced its greatest challenge since inception. I/O 2026 clearly reveals Google’s latest strategy: leveraging the world’s largest internet traffic gateway to natively embed AI across its entire ecosystem.
Search’s Biggest Overhaul in 25 Years: The Search Box Becomes an AI Gateway
One of the most pivotal changes at this year’s I/O is Google’s full AI-driven transformation of its search service.
According to reports, Google is redesigning its iconic search box to better handle users’ “longer, more complex questions” directed at AI chatbots.
The new search box will:
- Support more natural-language input;
- Enable easier image and file uploads;
- Automatically assist users in refining their queries;
- Support AI agents performing sustained tasks.
Nick Fox, Google’s Head of Knowledge and Information Platforms, said: “Users should be able to type any question they have in their mind directly into the search box.”
Google also plans to integrate Agent functionality into Search to help users: track topics of interest; book services; manage health information; and create personalized task dashboards. Some features will be available exclusively to paying subscribers.
Industry observers believe this signals Google Search’s evolution from a traditional “link distributor” into an AI-powered task gateway.
Gemini Omni: Google’s Most Advanced Multimodal Model Debuts
One of the most significant announcements at this year’s conference is Google’s unveiling of the new Gemini Omni model.
Multiple tech outlets reported that Gemini Omni is positioned as Google’s most advanced multimodal AI system to date—capable of simultaneously processing text, images, video, speech, and real-time interactive tasks.
Compared with earlier Gemini models, Omni’s biggest shifts are:
- Greater emphasis on real-time interaction;
- Stronger suitability for agent (intelligent agent) use cases;
- Cross-application and cross-device continuity in task execution.
Industry analysts view this move as Google’s direct response to OpenAI’s GPT-5 agent architecture and Microsoft’s advancing Copilot ecosystem.
Google executives repeatedly stressed at the conference: “AI isn’t just answering questions—it’s beginning to truly get work done.” This shift signifies Google’s strategic pivot—from an “AI assistant” to an “AI task-execution platform.”
Gemini 3.5 Flash Prioritizes Speed and Low Cost; Gemini 3.5 Pro Launches in June
Beyond its flagship models, Google also introduced Gemini 3.5 Flash.
According to I/O announcements and Google’s developer portal, the Flash series targets low-latency, high-responsiveness, and lower inference costs—optimized for developers, mobile devices, real-time interactions, AI programming, and automation scenarios.
As a lightweight addition to the Gemini family, Gemini 3.5 Flash delivers cutting-edge technical capabilities at half the price of comparable state-of-the-art models—and in some cases, as low as one-third. Google announced that Gemini 3.5 Flash will become the global default model for both the Gemini app and the “AI Mode” feature in Google Search.
Google stated it has enhanced Gemini 3.5 Flash’s cybersecurity safeguards, making it “less likely to generate harmful content while also reducing false rejections of safe queries.”
Google also disclosed that its more powerful variant, Gemini 3.5 Pro, is currently in internal use—but won’t roll out to broader users until next month, June.
Media reports note that Google clearly recognizes AI competition is no longer solely about “who’s the smartest,” but increasingly about “who’s faster, cheaper, and better suited for large-scale deployment.”
In the era of agents and real-time AI, latency and cost are rapidly gaining strategic importance.
Reports indicate Gemini 3.5 Flash will serve as one of the foundational models underpinning Google’s next-generation AI subscription framework.
Gemini Spark: Google Makes Its Big Bet on “AI Agents”
Beyond model upgrades, Gemini Spark drew the most market attention.
Google defines Spark as a “persistently running AI agent” capable of executing tasks continuously in the background. Multiple media outlets called this one of the most strategically significant launches at this year’s I/O.
According to Google’s demos, Gemini Spark will eventually be able to:
- Automatically organize emails;
- Monitor user calendars;
- Track web content;
- Manage documents;
- Execute cross-application workflows;
- Even perform select consumer actions and bookings on behalf of users.
Reports indicate Google is championing the concept of “always-on AI”—where AI no longer waits for user commands but proactively handles tasks.
Josh Woodward, Google Vice President, said: “Spark represents a major evolution for Gemini.” Industry observers believe this marks AI’s transition from the “chatbot era” into the “agent era.”
Previously, whether ChatGPT or Gemini, the core function remained fundamentally “conversational tools”; Spark, by contrast, moves much closer to a true digital assistant.
Clearly, Google aims to leverage its Gmail, Chrome, YouTube, Docs, and Android ecosystems to turn Gemini into the internet-level AI execution gateway.
“We Are Entering the Search Agent Era”
Google executives declared at I/O: “We are entering the search agent era.” Empowered by the Gemini 3.5 model, Google Search has achieved a performance leap—enabling complex task execution directly through search, such as planning travel itineraries.
Robby Stein, Google Search’s Product Vice President, took the stage to demonstrate Gemini’s intelligent agent capabilities. He said: “We firmly believe the best search experience is one that truly works for you.”
With Google Search’s new “AI Overviews” feature, users enjoy highly interactive experiences during searches—for example, deeply exploring the mysteries of black holes. For different search topics, the feature can even automatically generate corresponding animated demonstrations and customized user interfaces.
Google Search now possesses large-scale task-execution capability. It can autonomously build mini-apps to efficiently handle massive tasks. A typical use case: when users seek weekend family entertainment ideas, Google Search can generate a detailed, tailored weekend activity plan from scratch.
Google and Samsung’s AI Glasses Launch This Fall
Hardware was also a major theme at this year’s I/O.
Google publicly debuted its next-generation AI smart glasses for the first time—marking its return to the smart glasses market. Google announced partnerships with Samsung, eyewear manufacturers Gentle Monster and Warby Parker, to launch smart glasses powered by the Gemini assistant. Google stated these glasses will be compatible with both Android and iOS devices and are expected to go on sale later this year.
Shahram Izadi, Head of Android XR Products and Platforms, said: “These are just the first two designs in a broader lineup launching this fall.” He noted the glasses are designed to deliver information “privately into the user’s ear—not via screen display.”
Samsung stated the device will combine the Android XR platform with Gemini AI capabilities to deliver a “smart wearable computing experience.”
I/O demonstrations showed the AI glasses enabling: real-time translation; environmental recognition; navigation assistance; voice interaction; and access to Gemini agents.
Industry analysts believe Google is attempting to replicate Android’s ecosystem dominance from the smartphone era—integrating AI agents with AR devices to secure early leadership in the next computing platform.
Restructuring AI Subscriptions: Introducing $100/Month Developer Plan
Beyond models, another key change announced at the conference is Google’s overhaul of its AI pricing structure.
Google unveiled a new AI subscription plan for developers priced at $100 per month. This plan includes:
- Higher Gemini API call quotas;
- Advanced features in AI Studio;
- Gemini Code Assist;
- Google Cloud credits;
- Higher-tier agent access privileges.
Analysts interpret this as Google’s full-scale push into commercializing its developer offerings.
Previously, OpenAI and Anthropic had established first-mover advantages in enterprise and developer markets, whereas Google—despite possessing strong model capabilities—had progressed more slowly in cultivating its developer ecosystem.
Now, Google clearly aims to win back developers through more competitive pricing and deeper tool integration.
AI Ultra Monthly Fee Reduced from $250 to $200
Meanwhile, Google also adjusted its flagship AI subscription plan, AI Ultra.
AI Ultra’s monthly fee was previously $250; Google announced it will now be reduced to $200.
AI Ultra currently provides: access to the highest-tier Gemini models; Deep Think reasoning capability; advanced agent features; AI-powered video and image generation; AI programming tools; advanced NotebookLM features; and higher API and context-length limits.
Market observers believe this price cut reflects the rapid escalation of the AI subscription race.
Over the past year, OpenAI launched ChatGPT Pro; Anthropic strengthened Claude Max; Microsoft expanded Copilot enterprise subscriptions; and xAI advanced Grok Premium memberships. Google’s reduction of the Ultra price serves both to expand its user base and signals that large-model competition is shifting from a “technology race” to a “price war.”
Join TechFlow official community to stay tuned
Telegram:https://t.me/TechFlowDaily
X (Twitter):https://x.com/TechFlowPost
X (Twitter) EN:https://x.com/BlockFlow_News
