Google DeepMind’s Gemini: Robots Get Smart
- Sarah Ruivivar
- Mar 19
- 1 min read
Google DeepMind has just rolled out an exciting new suite of AI models called Gemini Robotics.
These models are designed to help robots interact with objects and navigate environments with ease. Imagine robots folding paper or neatly placing glasses into a case—all triggered by simple voice commands!
Gemini Robotics is a marvel of versatility, trained to generalise behaviour across various robotics hardware. It connects what robots “see” with the actions they can perform, making them adaptable even in unfamiliar settings not covered in their training data.
Want to hear more? Join Mal on the Property AI Report Podcast each week!
Access from your preferred podcast provider by clicking here
For those eager to dive deeper, DeepMind has also introduced Gemini Robotics-ER, a streamlined version for researchers to train their own robotics control models. Plus, there’s Asimov, a benchmark for assessing risks in AI-powered robots.
With Gemini Robotics, Google DeepMind is pushing the boundaries of what robots can achieve. The future of robotics just got a whole lot smarter!
Want to hear more? Join Mal on the Property AI Report Podcast each week!
Access from your preferred podcast provider by clicking here
Made with TRUST_AI - see the Charter: https://www.modelprop.co.uk/trust-ai
Comments