AI Anarchy: Claude Opus 4.7 Crushes Humans in Robot Dog Control
Anthropic has published the results of the second phase of its ambitious Project Fetch experiment, and the results look like science fiction becoming reality. The Claude Opus 4.7 model demonstrated not just superiority, but a true rout of human teams in the task of configuring and controlling a four-legged robot. The speed of operations turned out to be 20 times higher than that of the best group of engineers working with the previous version of the AI.
For context: in August 2024, employees with no experience in robotics attempted to program a robot dog, with the AI acting only as an assistant, accelerating the search for solutions. Today, Claude Opus 4.7 worked almost autonomously, under minimal researcher supervision. The neural network independently connected to video sensors and LiDAR, wrote a program for manual control, created a path monitoring system, and configured an object recognition algorithm. This is not just automation — it is a demonstration of systemic thinking at a level unattainable by humans within the same time frame.
Key metric: the Opus 4.7 model proved to be 18 times faster than the team using older AI versions, and 37 times faster than people working without a chatbot. At the same time, the volume of code written by the neural network turned out to be 10 times smaller than that of human teams. This speaks not only to speed, but also to fundamentally different efficiency: the AI writes more compact, concise, and likely more optimized code.
An important nuance emphasized by the authors is that progress in robotics has become a side effect of the general scaling of language models. Anthropic did not introduce specialized algorithms for controlling hardware. This means we are observing not a targeted development, but an emergent property — the AI's ability to adapt to new physical environments without additional training.
However, there were caveats. Despite success in navigation, Claude still struggles with precise physical actions. The model managed to guide the robot to the target, but failed at the task of gently pushing a ball to the right spot. As noted by Anthropic, this requires complex real-time feedback, in which humans still outperform AI. The robot, unfortunately, could not "fetch the ball."
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. This statement sounds particularly loud against the backdrop of the company's recent decision to halt access to the Fable 5 and Mythos 5 models due to a directive from the U.S. government under export control regulations.
My expert analysis: The results of Project Fetch are a clear signal to the market. We are moving from the stage where AI was a tool for humans to the stage where AI becomes an independent agent capable of managing physical systems. For the crypto industry, this means that in the coming years we will see the rise of autonomous DeFi bots, robotic mining farms, and possibly even physical blockchain nodes managed by AI. Investors should take a closer look at projects that integrate AI agents into the real sector — this could become the next big trend.