Google I/O Conference All in AI: New Gemini Model, Largest Revision of Search Box in 25 Years, Return of AI Glasses, Spark Smart Agent Debut

Original author: Li Dan

On the 19th of this month, Eastern Time, Google, a subsidiary of Alphabet, released a series of AI updates intensely at its annual developer conference Google I/O 2026, from redesigning Google Search to releasing the Gemini Omni and 3.5 Flash models as well as the Gemini Spark agents, and unveiling the new generation of AI glasses for the first time, Google is attempting to embed AI throughout its entire internet entry system—search, Android, office, video, and even next-generation hardware.

Sundar Pichai, CEO of Alphabet and Google, revealed in his keynote that the monthly active users of the Gemini app have doubled in a year to over 900 million; the “AI Overviews” feature in Google Search has now surpassed 2.5 billion monthly active users; and the number of users in the “AI mode” has exceeded 1 billion.

Pichai emphasized that AI is no longer just a chatbot but a “next-generation computing platform.” He stated at the conference, “AI is illuminating every part of the company.”

The Gemini Omni model, Gemini 3.5 Flash, and the Gemini Spark agents capable of executing tasks continuously, along with the restructuring of the AI subscription system, show that Google is trying to upgrade Gemini from a chatbot to an “AI operating system” that covers search, office, development, video, and mobile ecosystems.

This conference is also seen as a comprehensive counterattack by Google against competitors like OpenAI and Microsoft. Over the past year, as ChatGPT has changed user search and content consumption habits, Google's traditional search business has faced its greatest challenge since its inception, while the I/O conference clearly showcased its latest strategy: to use the world’s largest internet traffic entry to embed AI “natively” into the entire Google ecosystem.

The Biggest Revamp of Search in 25 Years: The Search Box Becomes an AI Entry Point

One of the core changes in this I/O is Google's complete AI transformation of its search business.

According to reports, Google is redesigning its iconic search box to better handle “longer, more complex questions” posed by users to the AI chatbot.

The new search box will:

Support more natural language input;
Be more convenient for uploading images and files;
Automatically assist users in refining their searches;
Support the continuous execution of tasks by AI agents.

Nick Fox, head of knowledge and information at Google, stated: “Users should be able to input any question from their minds directly into the search box.”

Google also plans to add Agent features to search, helping users: track topics of interest; book services; manage health information; create personalized task panels. Some features will only be available to paid subscription users.

Industry insiders believe this indicates that Google Search is shifting from a traditional “link distributor” to an AI task entry point.

Google's Strongest Multimodal Model Gemini Omni Debuts

One of the core updates at this conference is the introduction of the new Gemini Omni model.

Several tech media reports that Gemini Omni is positioned as Google’s most advanced multimodal AI system, capable of handling tasks involving text, images, video, voice, and real-time interaction simultaneously.

Compared to the previous Gemini series, Omni's biggest changes are:

Greater emphasis on real-time interaction;
More suitable for Agent scenarios;
Ability to continuously perform tasks across applications and devices.

Industry insiders believe that Google is directly targeting OpenAI's GPT-5 Agent system and the Copilot ecosystem being promoted by Microsoft.

Google executives emphasized repeatedly at the conference that “AI isn’t just answering questions; it’s starting to actually get work done.” This change indicates that Google's AI strategy has shifted from “chat assistant” to “task execution platform.”

Gemini 3.5 Flash Focuses on Speed, Low Cost 3.5 Pro Releases in June

In addition to the flagship model, Google also launched Gemini 3.5 Flash.

According to information from the I/O conference and Google's developer page, the Flash series is positioned for low latency, high response speed, and lower inference costs, focusing on: developers; mobile platforms; real-time interaction; AI programming and automation scenarios.

As a lightweight new member of the Gemini models, Gemini 3.5 Flash possesses cutting-edge technological capabilities, priced at only half that of similar advanced models, and in some cases as low as one-third. Google announced that Gemini 3.5 Flash will become the global default model for the “AI mode” in the Gemini app and Google Search features.

Google stated that it has strengthened the cybersecurity defenses of Gemini 3.5 Flash, making it “less likely to generate harmful content and harder to mistakenly reject safe query requests.”

Google also revealed that its more powerful version, Gemini 3.5 Pro, is currently for internal use, but will be released to a wider audience in June next month.

Media noted that Google clearly realizes that AI competition is no longer about “who is the smartest,” but rather who is faster, cheaper, and better suited for large-scale deployment.

In the Agent and real-time AI era, the importance of latency and cost is rapidly increasing.

Reports say that Gemini 3.5 Flash will become one of the important foundational models of Google's new generation AI subscription system.

Gemini Spark: Google Officially Bets on “AI Agents”

Compared to model updates, the market pays more attention to Gemini Spark.

Google defines it as a “continuously running AI agent” that can perform tasks in the background for a long time. Multiple media outlets describe this as one of the most strategically significant releases of this I/O.

According to Google's demonstration, Gemini Spark will be able to in the future:

Automatically organize emails;
Monitor user schedules;
Track web information;
Manage documents;
Complete workflows across applications;
Even represent users for some purchases and bookings.

Reports state that Google is driving a concept of “always-on AI,” where AI no longer waits for user commands but actively processes tasks.

Google Vice President Josh Woodward stated: “Spark represents a significant shift for Gemini.” Industry insiders believe this means the AI industry is transitioning from the “chatbot era” to the “agent era.”

Previously, whether ChatGPT or Gemini, the essence remained “dialogue tools”; whereas Spark is closer to a true digital assistant.

Google clearly hopes to leverage Gmail, Chrome, YouTube, Docs, and the Android ecosystem to transform Gemini into an AI execution entry point at the internet level.

“We Are Entering the Era of Search Agents”

Google executives stated at the I/O conference: “We are entering the era of search agents.” With the support of the Gemini 3.5 model, Google Search has achieved a performance leap, able to execute complex tasks like creating travel itineraries through search.

Robby Stein, the product vice president of Google Search, demonstrated the intelligent agent capabilities of Gemini. He stated: “We believe that the best search experience should be one that truly serves you.”

With the new Google Search “AI Overview” feature, users will obtain a highly interactive experience during their search—for example, deeply exploring the mysteries of black holes. For different search topics, the search function can even automatically generate corresponding animated demonstrations and customized user interfaces.

Google Search now possesses the capability to execute tasks on a large scale. It can automatically build mini-apps for users, aiding in the efficient processing of large-scale tasks. A typical application scenario is: when users want to find inspiration for family entertainment activities over the weekend, Google Search can start from scratch to tailor a detailed weekend activity planning proposal.

Google's Collaboration with Samsung on AI Glasses to Launch This Fall

Hardware is also an important theme of this I/O.

Google publicly unveiled its new generation of AI smart glasses for the first time, marking its return to the smart glasses market. Google announced that it has partnered with Samsung and eyewear manufacturers Gentle Monster and Warby Parker to launch smart glasses equipped with the Gemini assistant. Google stated that these glasses will be compatible with Android and iOS devices and are expected to be released later this year.

Shahram Izadi, head of Android XR products and platforms, stated: “These are just the first two designs in a larger series of products coming this fall.” He noted that the glasses are designed to provide information by “privately broadcasting to users’ ears rather than displaying on a screen.”

Samsung stated that this device will combine the Android XR platform with Gemini AI capabilities to create a “smart wearable computing experience.”

The display at the I/O conference showed that these AI glasses can: translate in real-time; recognize the surrounding environment; provide navigation; conduct voice interactions; and invoke Gemini agents.

Industry insiders believe that Google is attempting to replicate the ecological advantages of Android in the smartphone era, combining AI agents with AR devices to position itself in advance of the next generation of computing platforms.

Reconstructing AI Subscriptions: Introducing a $100/month Developer Plan

In addition to models, another important change at the conference is Google's redesign of the AI charging system.

Google announced a new AI subscription plan for developers, charging $100 per month. This plan will provide:

Higher quota for Gemini API calls;
AI Studio advanced features;
Gemini Code Assist;
Google Cloud credits;
Higher-level Agent access permissions.

Analysis suggests that this means Google is starting to fully push for developer commercialization.

Previously, OpenAI and Anthropic had established first-mover advantages in enterprise and developer markets, while Google, despite having powerful model capabilities, has been relatively slow in advancing its developer ecosystem.

Now, Google clearly wants to regain developers through lower prices and deeper tool integration.

AI Ultra Monthly Fee Reduced from $250 to $200

At the same time, Google has adjusted its flagship AI subscription plan AI Ultra.

Previously, the AI Ultra monthly fee was $250, but Google has announced that it will be reduced to $200.

AI Ultra currently offers: access to the highest level Gemini models; Deep Think inference capabilities; advanced agent features; AI video and image generation; AI programming capabilities; advanced functionalities of NotebookLM; and higher API and context limits.

The market believes that behind this price adjustment is the rapid escalation of the AI subscription war.

Over the past year, OpenAI launched ChatGPT Pro; Anthropic strengthened Claude Max; Microsoft expanded Copilot enterprise subscriptions; and xAI is also promoting Grok premium memberships. Google's reduction of the Ultra price is aimed not only at expanding its user base but also indicates that the competition for large models is shifting from “technology wars” to “price wars.”

免责声明：本文章仅代表作者个人观点，不代表本平台的立场和观点。本文章仅供信息分享，不构成对任何人的任何投资建议。用户与作者之间的任何争议，与本平台无关。如网页中刊载的文章或图片涉及侵权，请提供相关的权利证明和身份证明发送邮件到support@aicoin.com，本平台相关工作人员将会进行核查。