Nvidia Research Demonstrates Self-Trained Robots Using AI Coding Agents
Nvidia, Carnegie Mellon University, and UC Berkeley use AI coding agents to teach robots dexterous grasping with up to 99 percent success.

AI Coding Agents">
Researchers from Nvidia, Carnegie Mellon University, and UC Berkeley are using AI coding agents to teach robots dexterous grasping in the real world. A fleet of eight robots hits up to 99 percent success on tricky tasks. The approach leverages AI coding agents to enable robots to train themselves, which could significantly streamline the process of teaching robots to perform complex tasks.
By utilizing AI coding agents, the researchers aim to overcome the challenges associated with traditional robot training methods. The use of AI coding agents allows robots to learn from their environment and adapt to new situations, enabling them to develop dexterous grasping capabilities. This development has the potential to accelerate the deployment of robots in various industries, including manufacturing, logistics, and healthcare.
The success of this approach, with robots achieving up to 99 percent success on tricky tasks, underscores the potential of AI coding agents in robotics. As researchers continue to explore and refine this technology, it is likely to have a profound impact on the field of robotics. Why this matters: The ability to teach robots complex tasks through AI coding agents has far-reaching implications for industries that rely on automation.
By enabling robots to train themselves, businesses can reduce the time and cost associated with traditional training methods. This development also raises questions about the future of work, as robots become increasingly capable of performing tasks that were previously the exclusive domain of humans. As the technology continues to evolve, it will be crucial to address the potential societal implications and ensure that the benefits of automation are equitably distributed.
Source: The Decoder