Claude Opus 4.7 crushed humans in robot dog control: 20 times faster
Anthropic has unveiled updated results from the Project Fetch experiment, and they are impressive. The Claude Opus 4.7 model demonstrated the ability to configure and control a robot dog 20 times faster than teams of human engineers. This is not just an improvement — it is a paradigm shift in robotics.
Back in August 2024, employees with no robotics experience attempted to program a quadruped robot using AI that served only as an assistant. Today, the situation has changed dramatically. In the new testing phase, Claude Opus 4.7 worked almost autonomously, with minimal researcher supervision.
What did the neural network do independently?
- Connected to video sensors and LiDAR.
- Wrote a program for manual control.
- Created a system for monitoring the robot's path.
- Configured an object recognition algorithm.
The results speak for themselves: Opus 4.7 was 18 times faster than a team using older AI versions, and 37 times faster than humans working without chatbot assistance. Moreover, the neural network generated code that was 10 times smaller in volume than that produced by human teams. This means not only speed but also efficiency — less code, fewer errors, higher performance.
It is important to emphasize that progress in robotics has been a side effect of the general scaling of language models. Anthropic did not implement specialized algorithms for equipment control. This suggests that universal AI models are capable of adapting to new tasks without additional training.
However, there were limitations. Claude still struggles with precise physical actions. The model managed to guide the robot to a target but failed to gently push a ball to a specific point. This requires complex real-time feedback, where humans still hold an advantage.
Anthropic believes the industry is entering an era of "physical AI agents." In the future, neural networks will be able to use standard tools and equipment as effectively as they currently work with software code.
My comment as an analyst: This experiment clearly demonstrates that AI is transitioning from a purely digital realm into the physical world. The speed and efficiency of Claude Opus 4.7 are impressive, but the issue with precise manipulations reminds us that we are still far from full autonomy in the real world. Nevertheless, if the trend continues, we may see AI agents capable of managing complex equipment on par with humans — and possibly surpassing them — in the coming years.