A year after ChatGPT’s breakthrough, Google is joining the race. At the Mobile World Congress in Barcelona, Google unveiled Gemini Live, a development set to transform smartphones into interactive personal assistants capable of seeing and understanding the world around them in real time.
About a year ago, Google introduced Project Astra, a demonstration of AI capable of processing live video and answering questions in real time. Now, that technology is becoming a reality with Gemini Live, which will be available to Android users by the end of March.
Until now, Gemini could process text, images, and various types of documents. With Gemini Live, it gained a new ability: live video processing and screen sharing.
Simply put, users will be able to open their phone’s camera, show something to Gemini Live, and receive real-time answers and relevant information. Additionally, users can share their screens and ask for assistance with various tasks, such as online shopping or troubleshooting technical issues.
In a demonstration video, a user scanned their phone around a room. Gemini Live provided information about different objects, such as a computer code snippet, speakers, and a network diagram. The AI even remembered where the user had placed their glasses earlier in the video.
Gemini Live’s video-processing capabilities will roll out in the Gemini app for Android at the end of March, but access will be restricted to subscribers of the AI Premium plan, which costs $20 per month. This subscription grants access to Google’s most advanced AI models.
Processing live video is a computationally intensive task, requiring significantly more resources than text or image processing. Google acknowledges that running Gemini Live will demand greater processing power, which may result in substantial financial losses in the early stages.
However, the company believes that this innovation will drive greater adoption of Gemini and help it compete with OpenAI and other AI industry leaders.
Competing with ChatGPT’s “seeing eye”
Despite Google’s strong presence in the mobile market, Gemini’s adoption remains significantly lower than OpenAI’s AI tools, which have already been offering ChatGPT’s “seeing eye” feature for about a year. Google hopes that Gemini Live will close the gap and establish its AI as a serious contender in the race for the future of artificial intelligence.