Google Unveils Gemini: Advancing Multimedia Capabilities in AI with Bard Chatbot

Google’s Bard AI chatbot is advancing with a new model called Gemini, introducing native understanding of video, audio, and photos. Initially available in English through the Gemini update, Google Pixel 8 phone owners will benefit from enhanced artificial intelligence capabilities. Gemini offers text-based chat improvements for tasks like document summarization and programming code writing. The more significant change, incorporating multimedia abilities such as understanding hand gestures and interpreting drawings, will be rolled out soon.

Gemini represents a departure for AI, aiming to grasp richer information in our three-dimensional world, including complex communication abilities beyond written words. It comes in three versions tailored for different computing power levels:

1. Gemini Nano: Runs on mobile phones, powering features like conversation summarization and suggested message replies.
2. Gemini Pro: Tuned for fast responses, operates in Google’s data centers and powers the new Bard version.
3. Gemini Ultra: Limited to a test group for now, available in the Bard Advanced chatbot in early 2024.

This highlights the rapid advancement in generative AI, with Google’s Gemini being a response to competitors like OpenAI’s ChatGPT. Google aims to create AI that feels like a helpful collaborator and is evolving its models to align with human understanding.

Gemini’s training on text, programming code, images, audio, and video enables it to efficiently handle multimedia input. Examples of its abilities include recognizing patterns in shapes, linking photos to historical events, and processing complex physics problems. While multimedia integration awaits further testing, Google emphasizes responsible development and collaboration to address AI risks.