Anthropic's AI agent outperforms engineers by 20 times: Claude Opus 4.7 autonomously controls a robot dog
Anthropic's Claude Opus 4.7 model has made a breakthrough in physical AI, completing tasks to configure and control a robot dog 20 times faster than teams of human engineers. This was the result of the second phase of the Project Fetch experiment, where the AI operated almost autonomously.
How AI Surpassed Humans
In August 2024, company employees with no experience in robotics attempted to program a four-legged robot. At that time, the neural network only helped find solutions. In the new testing phase, Claude Opus 4.7 acted under minimal researcher supervision. The model 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 are impressive: Claude Opus 4.7 was 18 times faster than a team using older AI versions, and 37 times faster than humans without chatbot assistance. The key factor was that the neural network wrote more efficient code—its volume turned out to be 10 times smaller than that of human teams.
Evolution Without Specialization
An important point: Anthropic did not introduce specialized algorithms for controlling hardware. The progress in robotics was a side effect of the general scaling of language models. This suggests that fundamental AI development could lead to the emergence of agents capable of working with physical objects without additional training.
However, Claude still struggles with precise physical actions. The model managed to guide the robot to a target but failed at the task of gently pushing a ball to a specific point. This requires complex real-time feedback, in which humans still outperform AI.
A Look into the Future
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. Recall that on June 13, the company suspended access to the Fable 5 and Mythos 5 models due to a directive from the U.S. government under export controls.
My analysis: The results of Project Fetch are a clear signal that we are on the verge of a paradigm shift. When AI begins programming robots faster and more compactly than humans, the question is not whether machines will replace us, but how quickly we can adapt to new realities. For now, precise motor skills remain a weak point, but it is only a matter of time—and judging by the pace, not a very long one.