Google takes Gemini to another level: all the news of the most important AI event

He Google I/O 2025, The annual event for developers and technology enthusiasts celebrates its new edition, and in its inaugural day it has made it clear that The future of the company points to artificial intelligence. From the Shoreline Amphitheatre of Mountain ViewGoogle has shared a series of innovations focused on the integration of AI in all its products, in addition to presenting in extended reality, audiovisual creation and new subscription plans.

The Google CEO, Sundar pichaihe was in charge of welcoming attendees with enthusiasm, making clear the transformation that AI is generating in Google products and services. In addition, he stressed that “Gemini is now the core of our user experience” and? “AI is redefining the way we interact with technology.”

From Markeingdirecto.com – MKD We have been able to attend the annual Google conference to learn about all the news and technology trends presented during the event.

All the news of Google I/O 2025

With more than 2.5 million developers using Google’s tools, Pichai emphasized the company’s commitment to make AI accessible and useful for all. These are some of the novelties that Google presented in its event for developers:

Gemini 2.5 Pro: The center of Google’s new strategy

The great star of the event was Gemini 2.5 Prothe last evolution of Google’s artificial intelligence model. During the presentation there have been significant improvements in reasoning, multimodal understanding, code generation and security. Is now integrated directly into key services such as Gmail, Chrome y Google Search, allowing a more fluid, contextual and proactive interaction with the user. In addition, now Gemini can adapt to tone of voice, recognize emotions, and maintain more natural conversations thanks to functions such as affective dialogue and proactive audio.

Similarly, Google stressed that its infrastructure allows up to 480 billion token per month, Impressive growth compared to 9.7 billion the previous year. In addition, Gemini has proven to be the safest model against attacks by prompt injection, reinforcing the company’s commitment to the ethical use of AI.

Another outstanding fact is that The Gemini app already has more than 400 million monthly active users. «We observe strong growth and great interaction, especially with the models of the 2.5 series. For those who use version 2.5 PRO in the Gemini app, the use has increased by 45 %».

Project Astra, the universal assistant of Google

Another relevant ads is the presentation of «Project Astra»a Google Deepmind initiative that seeks to create a universal assistant with advanced observation, analysis and action abilities. Astra is able to understand environments in real time, search for information in emails, manuals or videos, and even make phone calls.

Its integration in Gemini 2.5 allows you to execute multiple complex tasks simultaneously, such as making reservations, sending messages or looking for specific information, all from a single artificial intelligence interface.

New ways to interact with Gemini Live and Canvas

Gemini Live It is another key novelty that allows Rich voice interactions, with camera entrance, shared screen and real -time content generation. It includes functions such as transforming drawings into shopping lists, generating infographics, creating audios in more than 45 languages ​​or developing applications from scratch.

In addition, tools such as Canvas and the new interactive questionnaires allow studying through personalized question games with immediate feedback, a functionality that was specially thought for students and educators.

Android XR and smart glasses

The company officially presented Android XR, its new platform for extended reality that will be launched in the late 2025 in collaboration with Samsung and Sony. This expected technology will subtitulate the world in real time, offer contextual information based on what we see and execute voice commands without the need to get the motive out of the pocket.

Android XR glasses will include integrated screen, contextual memory and simultaneous translation, which makes them a powerful tool for navigation, productivity and accessibility.

3D video call with Project Beam and real -time translation on Google Meet

Google also presented Project Beam, The evolution of its ancient Project Starline. This platform transforms 2D video calls into hyperrealist three -dimensional experiences, creating the feeling of being in the same room as our interlocutor. The system uses AI models to simulate depth, expressions and visual contact. The first tests will arrive this year thanks to an alliance with HP, initially in business environments.

On the other hand, complementing its commitment to connectivity without barriers, Google Meet will incorporate real -time translation between English and Spanish, with the future objective of adding more languages. This function will not only be responsible for translating the text, but also of conserving the original voice of the person, including intonation and nuances.

Flow, Image 4 and I see 3, the new new generation creative tools

In the field of content generation, Google presented Flow, a professional audiovisual creation platform even to the film sector. In addition, the models I see 3 and image 4 also presented that allow generating high quality videos and photographs, respecting detailed instructions and with visual coherence between scenes.

I see 3 stands out for its ability to generate scenes with realistic physical, personalized characters and native sound effects. Flow, meanwhile, includes advanced tools such as camera control, scenes edition, organization of elements and didactic examples in its flow TV gallery.

New subscriptions

With this new bet, new subscription models are also born, such as Google you have ultraa new IA subscription plan with the highest use limits and access to more powerful models and functions premium.

As explained from the company, Google ai Ultra is now available in the US for $ 249.99 per monthand will soon be available in more countries. Among the benefits of this subscription is the best version of the Gemini app, or anticipated access to the innovative I see model. In the coming weeks, Ultra subscribers will also have access to Deep Think in 2.5 Pro.

It will also be available Flowa new tool for the creation of films with AI is specifically designed for the most advanced models of Google Deepmind (see, image and gemini); Whiska solution that helps users explore and visualize new ideas quickly with text and images indications; access a NotebookLM with improved abilities; Gemini in Gmail, Docs, Vids and more; anticipated access to Gemini directly in Chrome; Access to the Agent Research Prototype «Project Mariner»an individual plan to YouTube Premium y 30 TB of storage.