AI Claude Opus 4.7 outperformed humans by 20 times in controlling a robot dog: a new stage in the evolution of autonomous agents
Anthropic has presented the results of the second stage of the Project Fetch experiment, and the results are impressive: the Claude Opus 4.7 model handled the setup and control of a four-legged robot 20 times faster than teams of human engineers working with previous versions of AI. This is not just a quantitative leap — it is a paradigm shift.
To recall, in August 2024, employees with no experience in robotics, using only AI, accelerated the search for solutions. Now, Opus 4.7 acted almost entirely autonomously, under minimal researcher supervision. The neural network independently completed a full cycle of tasks:
- connected to video sensors and LiDAR;
- wrote a program for manual robot control;
- created a trajectory monitoring system;
- configured an object recognition algorithm.
The key advantage is not only speed but also code quality. Opus 4.7 turned out to be 18 times faster than teams using older AI versions and 37 times faster than humans working without chatbot assistance. At the same time, the volume of written code was reduced by 10 times — the neural network generates more concise and efficient solutions.
It is important to emphasize: Anthropic did not introduce specialized algorithms for robot control. The progress in robotics became a side effect of the general scaling of language models. This suggests that fundamental improvements in AI architecture automatically extend to related fields.
However, full dominance is still far off. Claude still struggles with precise physical actions in real time. The model managed to guide the robot to the target but failed to gently push a ball — this requires complex sensorimotor feedback, where humans still maintain an advantage.
Anthropic believes the industry is entering an era of "physical AI agents." My analysis confirms: we are witnessing the convergence of language models and robotics. In the next 2-3 years, neural networks will begin working with standard hardware as effectively as they currently do with software code. This will change not only logistics and manufacturing but also the very structure of the labor market.
Expert opinion: The Opus 4.7 breakthrough is not just hype. If a model can accelerate development by 20 times without specialized training, then we are on the verge of exponential growth in autonomous systems. Investors should take a closer look at companies integrating AI into hardware. But do not forget: the physical world is more complex than code, and here humans are still irreplaceable.