Following OpenAI’s ChatGPT upgrades, Google has unveiled new features for its Gemini AI chatbot and the Circle to Search feature on smartphones.
The Gemini update for Android will now recognize the app you are using and your activities within it. For example, while watching a YouTube video, you can interact with Gemini for information about the video. The update also allows you to drag and drop images created by Gemini into Gmail, Google Messages, and other apps.
With the Gemini Advanced update, paid users can ask questions about PDF files without navigating through numerous pages. This feature will be available on hundreds of millions of devices in the coming months.
Gemini Nano, the compact AI model that operates natively on devices, is now multimodal, accepting text, images, and voice inputs. This improves features like TalkBack, aiding users with vision impairments by describing images and other content. Additionally, Gemini Nano can detect scam calls and provide real-time alerts when it identifies scam-like conversation patterns.
Google has also enhanced its Circle to Search feature to help students with their homework. When students highlight a problem, they will receive detailed steps to solve physics and math word problems directly within their digital materials.
Later this year, Circle to Search will be able to address more complex issues involving symbolic formulas, diagrams, and graphs, thanks to Google’s new LearnLM technology.
Currently, Circle to Search is operational on over 100 million devices, with plans to expand to an additional 100 million by year’s end.