The AI agent Claude Opus 4.7 outperforms humans by a head in controlling a robot dog: speed increased by 20 times
The Anthropic team has presented impressive results from the second phase of the Project Fetch experiment. This time, the Claude Opus 4.7 model demonstrated the ability to independently program and control a four-legged robot — and did so 20 times faster than the best teams of human engineers who worked on it last year.
A New Level of Autonomy
In August 2024, experiment participants with no robotics experience attempted to program a robot using an AI assistant. At that time, the neural network only helped find solutions. Now, the situation has changed dramatically: Claude Opus 4.7 worked almost autonomously 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 movement trajectory;
- configured an object recognition algorithm.
The performance comparison is impressive: Opus 4.7 turned out to be 18 times faster than the team using older AI versions, and 37 times faster than people working without chatbot assistance. Moreover, the neural network generated code that was 10 times more compact than human-written code — indicating high efficiency and optimization of algorithms.
A Side Effect of Scaling
It is important to emphasize: Anthropic did not implement specialized algorithms for controlling physical devices. According to the developers, the progress in robotics is a side effect of the general scaling of language models. This means we are observing an emergent property of AI — the ability to adapt to new tasks without specialized training.
Where AI Still Falls Short of Humans
However, not everything is so smooth. Despite success in navigation, Claude encountered difficulties when performing precise physical actions. The robot managed to reach the target but could not gently push the ball to the exact spot. This task requires complex real-time feedback — an area where humans still have the advantage.
At Anthropic, they believe we are entering an era of "physical AI agents." In the future, neural networks will be able to work with standard tools and equipment as effectively as they do today with software code.
My Analysis
This experiment confirms a key trend: AI agents are transitioning from a purely digital environment into the physical world. The speed and quality of task execution by Claude Opus 4.7 are impressive, but the limitations in precise motor skills remind us that full robot autonomy is still far off. Investors should pay attention to companies developing interfaces between AI and physical devices — this will become one of the main growth drivers in the next 2-3 years.
Let me remind you that on June 13, Anthropic suspended access to the Fable 5 and Mythos 5 models due to a directive from the U.S. government under export controls — which only underscores the strategic importance of these technologies.