Google has confirmed that its long-awaited Gemini AI feature is ready for launch. The company’s conversational Gemini Live will soon allow users to view live video and screen sharing, a feature previously demoed as Project Astra. This update enables users to show the robot something instead of telling it.
Currently, Google’s multimodal AI can process text, images, and various documents but struggles with accepting video inputs, sometimes summarizing YouTube videos and failing at other times. However, in March, the Gemini app on Android will receive a major update to its video functionality. Users will be able to open their camera or share their screen as a live video feed, allowing them to interact with Gemini Live.
The updated feature is part of Google’s focus on Gemini AI, which aims to improve human-AI interactions. A recent demo showcased how Gemini Live can answer questions in real-time as users interact with the AI. The system demonstrated its capabilities by explaining code on a computer screen and recalling user-specific information from earlier parts of the conversation.
This technology could be particularly beneficial for elderly individuals who may struggle with audio cues, but future generations growing up with AI/LLM interactions might find it more accessible. With this update, Google is moving forward in developing an AI system that can enhance people’s quality of life.
Source: https://arstechnica.com/google/2025/03/gemini-live-will-learn-to-peer-through-your-camera-lens-in-a-few-weeks