Currently, we are seeing more and more robotics and AI efforts in space. However, to scale human activity and energy in space, we need humans in space — and we need to enable them to control and interact with robotic systems remotely, support their cognitive and affective states, and using Brain-Computer Interfaces (BCIs) might be one of the solutions.
BCIs have the potential to fundamentally change how humans interact with spacecraft, robots, and other systems in space. By creating a direct interface between the human brain and machines, BCIs could reduce the limitations of traditional physical controls, enable faster and more intuitive interaction with remote robotic systems, and allow astronauts to operate complex machinery even in environments where conventional interfaces are impractical. BCIs in space can also go far beyond direct control of spacecraft or robotic objects. They could become a new human–machine interface layer for distributed AI infrastructure in space, enabling human–AI collaboration, cognitive-state monitoring, intelligent information selection, remote telepresence, support of optimal performance states, advanced astronaut training and simulation. BCI can support existing robotic and AI systems to understand human intent, attention, workload, and preferences, while potentially delivering relevant information and feedback back to the human. As human activity expands beyond Earth, BCIs could become an important layer of communication between people and increasingly autonomous systems operating across the Moon, Mars, orbital infrastructure, and eventually deeper space.
To make this possible, we need specialized compute in space that can efficiently process AI models and BCI signals while enabling low-latency communication between human brains, spacecraft, shuttles, orbital platforms, satellites, and, eventually, other planets. Such infrastructure could enable a distributed architecture in which human cognition, AI, robotics, and communications operate as a unified system across Earth and space.
We introduce the first-ever concept of an advanced AI accelerator designed to enable interstellar Brain-Computer Interfaces and AI models to run beyond Earth’s surface, helping enable a greater human presence in space.
Our payload consists of advanced, replaceable accelerator nodes with 6G RF links and direct laser communication capabilities, enabling high-bandwidth communication between planetary surfaces, spacecraft, orbital objects, and distributed space infrastructure.
All designed to be tested with the newly launched NASA’s Robotically Manipulated Payload Challenge, specifically In-space Servicing, Assembly, and Manufacturing (ISAM).
That's one small step for a man, one giant leap for the brains.
We thank our collaborators for continuous support and scientific partnerships:
Eugene Hauptmann, Eduardo Baena, Gary Strangman, Vladimir Ivkovic, Schuyler Johnson, Njeri Gachoka, Sebastian Monzón - all from MIT, Harvard University, Northeastern University, Mass General Brigham, UC Santa Cruz and University of Cincinnati.