Google made some exciting announcements on day one of Google I/O 2024. The event’s highlight was the capabilities of Google’s AI model Gemini and its integration across other Google services. Google is also working on new AI-based image, video, and music generation tools. Here is all the new stuff from Google I/O 2024.
Long Context Conversations in Gemini
Google is now expanding the availability of Gemini 1.5 Pro to all users and developers. This powerful AI model supports up to 1 million tokens, allowing users to have longer contextual conversations. Google says that Gemini can now scan up to 1500 pages of documents and provide a summary within minutes.
The company is also working to bring audio conversation support to the official Gemini app on Android. Users will soon be able to have real-time voice conversations with the chatbot. Google also showcased a glimpse of its project Astra, which will be integrated into Gemini later.
Astra integration in Gemini will allow users to open their phone’s camera and scan any item in their surroundings. Gemini will analyze the video feed in real-time and explain it, and users can also ask questions about it.
Gemini can also be customized for specific tasks using Gems. This feature creates a profile in Gemini for a specific use. For example, users can set a Gem to help them train with their gym workouts, create a new recipe, plan a trip itinerary, and other specific tasks.
Gemini Integration Across Google Workspace
Google is now extensively deploying the power of Gemini AI across its Workspace applications, such as Sheets, Documents, Drive, Gmail, and other services. Users can now ask Gemini to summarize an entire email thread in Gmail. They can also ask the chatbot to search for certain emails with a specific context.
Gemini can now also access data from Google Sheets. Users can ask the chatbot questions to search for certain datasets. For example, users can ask Gemini to search for the most expensive item in the sheet, and the chatbot will respond accordingly.
Generative AI Media Tools
Google has announced new tools that can generate images, videos, and music using AI. The company has unveiled Imagen 3, a powerful tool that uses AI to generate high-quality images from textual prompts. It also lets users edit the image using preset brushes and other photo editing tools.
Veo is a tool for generating 1080p videos using textual prompts. It combines the potential of all previous video-generating tools by Google, such as Videpoet.
The company is also working with music creators to help them create new sound effects and even entire music tracks. This feature will be available to musicians and singers through the Music AI Soundbox.
LearnLM Tools for Student Learning
Google also unveiled LearnLM, a new set of tools designed to help students learn using generative AI. This feature is best used with educational videos on YouTube, where students can raise their hands to ask any question while watching academic content.
LearnLM can generate a summary of longer educational videos on YouTube. It also provides an option to quiz the students based on the contents of the video. Google says that this will introduce interactive learning and improve curiosity in the minds of students.
The new features announced at Google I/O 2024 are slowly rolling out to users, starting with the United States. An exact country-wise rollout timeline has not been shared yet. We can expect all features to be available across countries by the end of this year.
