CIOInsider India Magazine

Google's largest developer conference, I/O, features product announcements from all areas of the company's business. Media houses are already busy swarming around YouTube, Google Search, Android, Chrome, and, of course, Google's AI-powered chatbot, Gemini. The search giant revealed a number of new AI technologies that it has been developing over the past few months, including an asynchronous AI coding agent, an AI-first 3D video collaboration platform, and an AI tool for filmmaking. Here is a rundown of the event’s major launches that are taking place currently.

Gemini Ultra
Google says that Gemini Ultra, which is now available exclusively in the US, offers the "highest level of access" to its AI-powered apps and services. This indicates higher constraints apply to AI Ultra in Google's NotebookLM platform and its picture remixing tool, Whisk. In addition, Google's Gemini chatbot in Chrome, YouTube Premium, a few "agentic" tools utilizing the company's Project Mariner technology, and 30TB of storage across Google Drive, Google Photos, and Gmail are all available to AI Ultra users.

Veo 3
According to Google, Veo 3 can produce dialogue, sound effects, and background noise to go with the videos it produces. According to Google, Veo 3 also outperforms Veo 2 in terms of the caliber of video it can produce.

Veo 3 can be triggered with text or an image and will be accessible in Google's Gemini chatbot app for users of Google's $249.99/month AI Ultra plan.

Imagen 4
Imagen 4 surpasses Imagen 3 in speed, according to Google. Google plans to produce an Imagen 4 variation soon that is up to ten times faster than Imagen 3. Google claims that Imagen 4 can display "fine details" such as textiles, water droplets, and animal fur. It can produce graphics with up to 2K resolution and a variety of aspect ratios, handling both photo realistic and abstract styles. Flow, the company's AI-driven video tool designed for filmmakers, will be powered by both Veo 3 and Imagen 4.

Gemini App Updates
More than 400 people use Gemini apps each month, according to Google. This week, all iOS and Android users will be able to enjoy Gemini Live's camera and screen-sharing features. The Project Astra-powered capability allows users to feed video from their smartphone's camera or screen to the AI model and hold near-real-time conversational discussions with Gemini.

In the upcoming weeks, Gemini Live will also begin to further connect with its other apps, according to Google. It will soon be able to generate to-do lists using Google Tasks, offer directions from Google Maps, and create events in Google Calendar.

Stitch
Stitch is an AI-powered tool that generates the required UI elements and code to assist users in designing the front ends of web and mobile applications. With just a few words or even an image, Stitch may be asked to construct app user interfaces. It will then provide HTML and CSS code for the designs it creates. Compared to some other feel coding tools, Stitch's functionality is a little more constrained, but there are still plenty of customizing choices.

Additionally, Google has made Jules, their AI agent designed to assist developers in resolving coding errors, more widely available. In addition to handling specific backlog items and programming jobs, the tool assists developers in comprehending complex code and creating pull requests on GitHub.

Project Mariner
Google's experimental AI browser and user is called Project Mariner. Google claims to have made major updates to Project Mariner's functionality, enabling the agent to do almost a dozen activities concurrently, and is currently releasing it to consumers. Users of Project Mariner, for instance, can buy groceries or sports tickets online without ever going to a third-party website. Google's AI agent can be contacted via chat, and it will visit websites and perform things on their behalf.

Gemini AI
The Gemini AI app, a variety of new experiences in Search, and products from outside developers will all be powered by Google's low latency, multimodal AI experience, Project Astra. Google DeepMind created Project Astra to demonstrate multimodal, almost real-time AI capabilities. The company claims that it is now working with partners like Samsung and Warby Parker to produce those Project Astra spectacles, but it has not yet announced a specific launch date.

Project Astra
The Gemini AI app, a variety of new experiences in Search, and products from outside developers will all be powered by Google's low latency, multimodal AI experience, Project Astra. Google DeepMind created Project Astra to demonstrate multimodal, almost real-time AI capabilities. The company claims that it is now working with partners like Samsung and Warby Parker to produce those Project Astra spectacles, but it has not yet announced a specific launch date.

AI Mode
Google is introducing AI Mode, an experimental Google Search function that enables users to ask intricate, multi-part queries using an AI interface, to users in the US. AI Mode will provide "try it on" alternatives for clothing and enable the usage of sophisticated data in financial and sports searches. Later this summer, Search Live will launch, allowing you to ask inquiries based on real-time images captured by your phone's camera.

Beam 3D Teleconferencing
Beam, formerly known as Starline, allows users to communicate with others as if they were in the same conference room by utilizing a combination of hardware and software, such as a six-camera array and a unique light field display. The cameras are aimed at the user and positioned at various angles; the video is converted into a 3D depiction by an AI model.

Google's Beam touts 60 frames per second video streaming and "near-perfect" millimeter-level head tracking. Beam's AI-powered real-time speech translation capability, when used with Google Meet, maintains the original speaker's voice, tone, and facial expressions.

Also Read: Microsoft's 50th Anniversary: Highlights, Brave Steps and Road Ahead

Wear OS 6
For a better app appearance, Wear OS 6 unifies fonts across tiles, and Pixel Watches will gain dynamic theming that matches watch faces with app colors. The new design reference platform's main goal is to enable developers to create more individualized apps with smooth transitions. Figma design files and a design guideline for developers are being released by the company.

Google Play Store
With new tools to manage subscriptions, topic pages that allow users to delve into certain interests, audio samples that give consumers a preview of the app's content, and a revamped checkout process to streamline the add-on sales process, Google is strengthening the Play Store for Android developers. Users can access apps linked to a plethora of series and movies through "topic browse" sections for movies and television shows (currently limited to the US). Additionally, developers receive tools to monitor and enhance their app rollouts, as well as specific pages for testing and releases. If a serious issue arises, Google developers can now stop live app releases.