Google’s annual developer conference, Google I/O 2025, was held on May 20-21, 2025, at the Shoreline Amphitheatre in Mountain View. At this event, Google unveiled its best tech, including Gemini, AI Search, smart glasses, and more. CEO Sundar Pichai said that the company is now not taking time to launch its best tech. They are, in fact, launching robust AI features as soon as they’re ready.
Among several new developments, Gemini stole the show. Google launched the newest version of Gemini, Gemini 2.5 Pro. It is said to be one of the top-performing models out there. Aside from that, the tech giant has also launched a lighter, faster version of it called Gemini 2.5 Flash. Gemini 2.5 Flash is going to be a great companion for everyday tasks. The company made a lot of other announcements at the event, which we have detailed here in this post. So, reading it….
Table of Content
- Major Announcements Google Made at I/O 2025
- Gemini 2.5 – The Star of the Show
- Gemini in Gmail
- Gemini 2.5 Flash will be Available in June
- Gemini 2.5 Flash and Pro include Native Audio Outputs
- Google Announces a New AI Coding Agent
- Google Announced a Browsed-Based Agentic AI
- Google Announced Two New AI Subscription Plans
- Google Announced New Versions of Generative Media Models (Veo, Flow, and Imagen)
- Searching Made Easier by Google
- The Gmail AI Tools That You May Need
- Google Announced Project Starline as Google Beam
- Google Partnering with HP
- Google Expands AI Overviews
- Android XR on Smart Glasses
- Google Partners with Gentle Monster and Warby Parker
Major Announcements Google Made at I/O 2025
Here is the rundown of all the major announcements that Google made at the annual Developer Conference 2025. Take a look…
1. Gemini 2.5 – The Star of the Show
Google has launched the latest version of Gemini, Gemini 2-5. This multimodal AI assistant will be better, faster, smarter, and more theoretical. Google released Gemini 2.5 ahead of schedule. The advanced AI model was originally scheduled to be released at I/O; however, the brand launched weeks ago. This is due to the positive response from beta testers. The tech giant announced that the Gemini 2.5 Pro and Flash models will have several enhancements, including Deep Think support for the former. This enables the AI to employ advanced research methods to evaluate various hypotheses prior to delivering a response.
Speaking about Deep Think, DeepMind’s Demis Hassabis said, “Deep Think uses our cutting-edge research in thinking and reasoning, including parallel techniques.” It is currently being rolled out to Google’s trusted testers.
According to the tech giant, Gemini 2.5 Flash has also been enhanced in various aspects, including reasoning, multimodality, coding, and response efficiency, necessitating 20% to 30% fewer tokens, based on the company’s assessments.
Additionally, the Gemini 2.5 models will also support audio-visual input and native audio output dialogue through a preview version in the Live API. This enhancement enables developers to create and refine the tone, accent, and speaking style of conversational experiences.
According to Google, Gemini 2.5 Pro is an AI that’s “intelligent, understands the context you are in, and can plan and take action on your behalf across any device.”
2. Gemini in Gmail
Google has also launched personalized smart replies in Gmail. It is to help users instantly draft emails that match their specific context and tone. The feature will be available in Gmail this summer for subscribers. It takes ideas from the user’s past emails and Google Drive and accordingly provides suggestions that are more appropriate to the user.
3. Gemini 2.5 Flash will be Available in June
An updated version of Gemini 2.5 Flash will be available in June as the brand will be making final adjustments based on the feedback.
4. Gemini 2.5 Flash and Pro include Native Audio Outputs
Google is set to include a preview version of audio-visual input and native audio output dialogue for both Gemini 2.5 Flash and Pro. It will help users to guide the AI-generated tone, accent, and style of speech. Also, it can switch between languages (for instance, from English to Hindi) mid-speech.
5. Google Announces a New AI Coding Agent
The tech giant has also announced a new AI coding agent called Jules. Powered by Gemini 2.5 Pro, Jules is capable of autonomously reading and generating code. Users would be able to integrate this new AI agent directly into their existing code repositories. Jules can then make a clone of the user’s codebase within the Google Cloud virtual machine (VM) to ‘understand’ the context of the project and perform tasks accordingly. Some of the tasks that it can perform include writing tests, building new features, fixing bugs, etc.
6. Google Announced a Browsed-Based Agentic AI
Google has also unveiled Project Mariner. It is a browser-based agentic AI. It is capable of handling 10 different tasks simultaneously. With Agentic AI, you will not have to wait to do tasks at different times. It will help you manage things at the same time. You can use this to book flights, do research, and shop. The latest version of the research prototype will first be made available to Google AI Ultra subscribers in the US.
7. Google Announced Two New AI Subscription Plans
Google has announced two new subscription plans. According to the company, AI Pro will now be available with a full suite of AI products, higher limits, and Gemini Advanced. It is priced at $19.99/month. This includes the Gemini app (with Advanced), Flow, NotebookLM, and more.
While Google AI Ultra is priced at $249.99/month AI subscription plan. It will allow users to have higher usage limits and early access to its top-tier AI models and features, such as upcoming experimental AI products like Project Mariner, Veo 3, and Gemini 2.5 Pro with Deep Think mode. AI Ultra will only be available to users in the US for now.
The subscription, priced at $250 per month (or a 50% discount for the first three months for new users), includes early access to Agent Mode. This feature allows you to create prompts on a desktop level, enabling live web browsing, research, and additional functionalities, all managed by Google’s AI models.
Also See: 10 Best Smart Alarm Clocks With Alexa or Google Assistant
8. Google Announced New Versions of Generative Media Models (Veo, Flow, and Imagen)
Google has also released new versions of its generative media models, including Veo 3, Imagen 4, and Flow. Veo 3, Imagen 4, and Flow are next-level creator tools.
The new version of Veo 3 will now support audio prompt generation, such as traffic noises in a busy street, birds singing in a park, or even dialogue between characters. Earlier, it could only visualize prompts, with a muted playback, which made the content look dull and mundane in comparison to real-life videos.
Another tool that Google has announced at I/O 2025 is Flow. Flow is an AI filmmaking tool. It enables creators to adjust the way videos are produced. It helps creators with everything, from adjusting camera angles and motions to the cast and location to enhance the video output.
Imagen 4 is yet another generative media model launched at I/O 2025. It will help creators get accurate and clear details, particularly the intricate ones, such as fabric textures, water droplets, and animal fur. This AI image generator can create content in multiple aspect ratios. Also, it supports resolutions of up to 2K.
9. Searching Made Easier by Google
Google rolled out AI Mode in Search earlier this year. The feature will now be available in the US without Labs sign-up and with some new capabilities. Google has introduced Deep Search in AI Mode. The new feature will expand the number of background queries from tens to hundreds to deliver more robust and thought-out search responses. Users will be able to get a fully cited report, which, according to Google, will take just minutes to research and create.
Aside from that, the Google Search experience will also have Project Astra’s multimodal capabilities for visual assistance. Users can effortlessly direct their camera toward an object or scene and enquire about it in the same manner as they would with a Google image search.
Then, there is a new AI Mode for enhanced shopping experience. It will help users find inspiration, narrow down buying options, and even see themselves in an outfit that they wish to buy or are considering (via an image generator). All you will have to do is upload an image of yourself.
AI Mode will also include Project Mariner’s agentic functions. It will allow users to prompt Google to find the best event ticket deals or book restaurant appointments. The AI will present multiple options, including the one that best fits your inquiry to help you make a mindful choice.
10. The Gmail AI Tools That You May Need
Google’s AI integration reaches its Workspace services, encompassing Gmail, Meet, and Docs. Gmail will now have Gemini for personalized smart replies. The feature will, however, be available to subscribers. It will provide relevant responses based on your past replies to a contact or thread. It will provide a response according to tone and context, whether it’s formal or more casual and conversational.
Additionally, there will be a new inbox cleanup feature that you can use to prompt Gemini to delete emails from a specific sender within a given time frame.
Furthermore, Gemini will now propose a meeting booking window if it identifies that you’re attempting to arrange a meeting in a Gmail thread, eliminating the need for you to open your Calendars or Meet.
Google Meet will also include near-real-time speech translation. It will translate spoken words to the listener’s preferred language. The beta feature will first be rolled out to Google AI Pro and Ultra subscribers.
Finally, Users can now create source-based writing assistance in Docs. This will allow users to set specific source links from which Gemini will exclusively draw information and suggestions. The feature is set to be generally available in the upcoming quarter.
11. Google Announced Project Starline as Google Beam
Google has now branded Project Starline as Google Beam. Google Beam will now use AI to transform standard 2D videos into more realistic 3D experiences. It is similar to the multi-layering technology that is there in Apple and Meta’s spatial video processing. It is to allow users to build trust and a deeper understanding through realistic body movements, cues, and eye contact.
12. Google Partners with HP
Google also announced that it is moving one step ahead by partnering with HP. In its efforts to expand into the enterprise sector, Google is collaborating with HP, which has recently acquired Humane for its expertise in AI hardware, to develop communication devices powered by Google Beam.
13. Google Expands AI Overviews
Google has also expanded AI Overviews to over 200 countries and territories, as well as more than 40 languages, including Arabic, Chinese, Malay, Urdu, and more.
14. Android XR on Smart Glasses
Google also focused on Android XR glasses, practical wearables designed to utilize cameras, microphones, and speakers to enhance your visual experience and support Gemini. During I/O, Google showcased the Gemini capabilities of Android XR, featuring a configurable in-lens display that allows users to access a directional system for navigating city streets, akin to the HUD screens found in certain vehicles.
It will allow users to visualize and respond to incoming text messages, translate conversations in real-time, and capture photos using voice commands. This will significantly reduce the time and effort required for these tasks by saving you from the hassle of retrieving your phone from your pocket.
15. Google Partners with Gentle Monster and Warby Parker
Lastly, Google also announced that it is collaborating with renowned eyewear brands such as Gentle Monster and Warby Parker to develop more fashionable smart glasses. It is to reach a broader audience. According to the tech giant, the Android XR glasses will be released later this year.
The Bottom Line
So, these are the major announcements that Google made at I/O 2025. Hopefully, it has been an enlightening post for you. Thanks for reading! Stay tuned for more such insightful articles!
