
Google has announced Project Suncatcher, an experimental initiative aimed at testing machine learning infrastructure in low Earth orbit. On October 1, 2026, a SpaceX Falcon 9 rocket will launch the MVP satellite, which carries four of Google's custom Tensor Processing Units (TPUs). The project seeks to determine if space-based data centers can leverage solar energy for scalable AI operations. Engineers have successfully tested the hardware against extreme launch vibrations and radiation levels, confirming the TPUs can withstand orbital conditions. A significant technical hurdle remains: cooling. Due to the lack of atmospheric airflow in the vacuum of space, the chips must operate in fifteen-minute intervals to prevent overheating. Google is collaborating with Planet Labs to accelerate this research, with plans to test inter-satellite laser communication in 2027. This moonshot project represents a critical step toward potentially offloading massive AI workloads to orbital infrastructure.
This is a summary. Read the full article at the original source:
TechRadarRelated stories
The article examines the topic of inbreeding, which is traditionally viewed through the lens of biological 'protection' against degeneration. The auth…
Consciousness as a property of living matter to organize information about its own structure
In this article, the author proposes a new perspective on the nature of consciousness, moving away from traditional computational models that view the…
Caroline M. Yoachim’s short story, 'Welcome to the Medical Clinic at the Interplanetary Relay Station,' offers a poignant look at the intersection of…



