DeepMind Announces Plans to Utilize AI Models for Enhancing Physical Robots
Google DeepMind’s New AI Models for Robotics
Introduction to Gemini 2.0
On Wednesday, Google DeepMind unveiled two exciting AI models designed specifically for robotics operations. These models, named Gemini Robotics and Gemini Robotics-ER (Extended Reasoning), are powered by Gemini 2.0, which Google hails as its most advanced AI to date. Unlike previous generative AI that primarily handled text and image outputs, Gemini Robotics extends its capabilities to execute physical actions, enabling robots to perform real-world tasks.
A Partnership with Apptronik
To bring Gemini 2.0 to life, Google announced a collaboration with Apptronik, a robotics firm based in Texas. This partnership aims to develop the next generation of humanoid robots utilizing these new AI models. Apptronik has established itself as a significant player in the robotics sector, having previously worked with notable organizations like Nvidia and NASA. Notably, Google has also participated in Apptronik’s recent funding round, which raised $350 million.
Showcasing the AI Capabilities
Google provided a glimpse into the potential of its AI models through demonstration videos featuring Apptronik robots. These robots performed various tasks, such as plugging devices into power strips, filling lunchboxes, and organizing plastic vegetables—all in response to spoken instructions. However, Google has not disclosed any timeline regarding when these robotic technologies will be introduced to the market.
Key Qualities of Effective Robotic AI
In a blog post, Google outlined three essential qualities that AI models for robotics should possess to truly assist people:
- General Adaptability: The AI should be capable of adjusting to various situations and environments.
- Interactivity: It must understand and respond promptly to instructions and changes around it.
- Dexterity: The AI should replicate the fine motor skills humans use, allowing it to manipulate objects with precision.
The Role of Gemini Robotics-ER
Gemini Robotics-ER serves as a foundational tool for roboticists looking to train their own AI models. The model is accessible to Apptronik and selected "trusted testers," including Agile Robots, Boston Dynamics, and Enchanted Tools, enabling a range of developers to push the boundaries of robotic capabilities.
The Competitive Landscape in Robotics AI
Google isn’t the only major player in the fusion of AI and robotics. In November, OpenAI invested in Physical Intelligence, a startup focused on integrating general-purpose AI into physical environments. This company is dedicated to creating comprehensive AI models and algorithms that can effectively power robots.
OpenAI also made news by hiring the former head of Meta’s Orion initiative to spearhead its robotics and consumer hardware division. Tesla has also entered the humanoid robotics market with its Optimus robot, showcasing the increasing interest in this field.
Google’s Robotics Vision
Sundar Pichai, Google’s CEO, expressed that the company views robotics as a vital testing ground for implementing AI advancements in the physical realm. He highlighted the adaptability of the robots, which will utilize Google’s multimodal AI models to make real-time adjustments based on their surroundings.
The Future of Robotics
As the demand for sophisticated robotic systems continues to grow, the innovations from companies like Google and its partners signal a promising direction. The combination of advanced AI and robotics has the potential to transform industries, making tasks more efficient and revolutionizing how we interact with technology in our daily lives.
These advancements indicate a significant evolution in AI and robotics, paving the way for a future where these technologies might play an even greater role in everyday life.
