Google is fully committed to AI technology, and this was made clear during their recent keynote at the I/O developer conference, repeatedly emphasizing AI over 120 times. Despite frequent mentions, not every AI development announced by Google was groundbreaking; many of them were simply small steps forward or upgrades of existing technologies. However, Google I/O 2024 featured some outstanding AI products and features that highlight Google’s innovation efforts.
Generative Artificial Intelligence in Search
Google intends to revolutionize the way Google search results are organized using generative artificial intelligence. Depending on what you’re looking for, results can include AI-generated summaries of reviews or discussions from platforms like Reddit, and lists of recommendations. Initially, these enhanced search results pages will appear when users are looking for inspiration, such as when planning a trip, and will later be expanded to include searches related to restaurants, recipes, and more, covering a wide range of topics such as movies, books, and online stores.
Astra and Gemini Live Project
Improvements to Google’s artificial intelligence chatbot Gemini include a new feature called Gemini Live. This upcoming feature enables deep voice interaction with AI on mobile devices, where users can interrupt and ask questions, and the AI will adapt to their speaking style and respond to visual inputs from the user’s environment through the camera. Gemini Live, which is scheduled to launch later this year, is supported by the Astra project, which aims to develop multi-functional real-time artificial intelligence applications.
Google Veo
To surpass OpenAI’s Sora, Google introduced Veo, an artificial intelligence model capable of generating high-quality videos of up to a minute based on text prompts. Veo understands a variety of visual styles and can edit the footage to display the desired cinematic effects with simple command prompts. The model can also create longer story videos from a series of prompts, offering creators new storytelling tools.
Ask for Photos
An interesting addition to Google Photos is the Ask Photos feature, which is powered by Gemini AI models. Launched over the summer, the feature provides more intuitive natural language searches of users’ photo libraries. This will allow users to perform broad searches and AI will analyze different elements to get the best images based on quality and contextual data.
Gemini in Gmail
Gmail will soon include Gemini to help with tasks like searching, summarizing, and composing emails, as well as managing more complex actions like handling returns. At I/O, Gemini was shown summarizing school-related emails for parents, proving its ability to handle complex tasks within Gmail.
Scam Detection During Calls
An upcoming feature for Android will notify users of potential fraud during phone calls using the Gemini Nano. This scaled-down version of Google’s artificial intelligence will run entirely on the device, detecting patterns in conversation related to fraud. While this feature will be optional, it demonstrates the shift towards using AI to improve user security without compromising privacy.
Ai For Accessibility
Google is also focusing on accessibility, improving its TalkBack feature with AI capabilities. Gemini Nano will soon help provide audio descriptions of objects to help visually impaired users. This will help identify and describe images that are often left unlabeled, thereby enriching the interaction with TalkBack users.