Space-based data centers and AI sovereignty
MIT Science Policy Review · 07.08.2026 · 7 · 112-120
DOI: 10.38105/spr.jsnuhwszmu ↗Kurz gefasst
Fachartikel in der Zeitschrift MIT Science Policy Review, online erschienen am 07.08.2026. Verfasst von Carissma McGee.
Abstract (Originalsprache)
The notion of AI sovereignty has become an increasingly pertinent objective for nation-states and private companies developing, deploying, and operating systems critical to national security. AI sovereignty has become a prerequisite across product development life cycles. In this paper it is understood as a layered set of resources required to build and operate AI systems [1, 2]. Because these resources currently depend on infrastructure that can be geographically located and licensed, proposals to place data centers in orbit challenge sovereignty concepts that are typically grounded in territorial jurisdiction [3]. This paper explores the implications of pursuing AI sovereignty through the development of space-based data centers. In particular, it highlights the potential geopolitical challenges and risks posed by the pursuit of AI sovereignty in the space domain by examining defining geopolitical events that have shaped policies, technologies, and practices in the space sector: the Cold War space race, Europe’s development of its own Global Navigation Satellite System (GNSS), the formation of International Traffic in Arms Regulations (ITAR), and the U.S. Wolf Amendment of 2011. These events serve as analogies for examining how geopolitics and technology have shaped access, dependence, and concentration of sovereignty in the 21st century [4].
Zitieren
APA
McGee, C. (2026). Space-based data centers and AI sovereignty. MIT Science Policy Review, 7, 112-120. https://doi.org/10.38105/spr.jsnuhwszmu
Vancouver
McGee C. Space-based data centers and AI sovereignty. MIT Science Policy Review. 2026;7:112-120. doi:10.38105/spr.jsnuhwszmu
BibTeX
@article{McGee2026,
author = {McGee, Carissma},
title = {Space-based data centers and AI sovereignty},
journal = {MIT Science Policy Review},
year = {2026},
volume = {7},
pages = {112--120},
doi = {10.38105/spr.jsnuhwszmu}
}
EUVIMED