The emergence of China’s DeepSeek has significantly impacted the artificial intelligence sector, creating exciting new possibilities for space companies exploring the use of AI in space. AI is crucial for achieving true orbital autonomy and managing the increasingly crowded space domain. While companies like Loft Orbital are integrating AI, widespread industry adoption is still in its early stages.
DeepSeek boasts high performance with substantially lower computational requirements than other generative AI models. This efficiency is vital for space applications due to limited bandwidth and onboard processing power. Its open-source nature further accelerates broader AI innovation. If DeepSeek can run efficiently on edge devices or within smaller models, real-time AI decision-making becomes more practical for autonomous satellites, deep-space exploration, and resource-constrained environments.
Douglas Marsh, CEO of Martian Sky Industries, a company utilizing AI for orbital debris mitigation technology, believes DeepSeek’s innovations could propel large language models into space, boosting autonomy. “If you can get systems that are talking to each other — that are trained and have some sort of wrapper around them that’s essentially defining what they’re looking at, what they do, and what they process — you can begin to further automate systems,” Marsh stated at the SpaceCom conference on January 30.
Isaac Passmore, technologist fellow at ASRC Federal, highlights that while DeepSeek's capabilities aren't superior to other generative AI platforms, its rapid training speed is a major breakthrough. “It means that you can solve problems faster,” Passmore noted during a SpaceCom panel. He anticipates more frequent disruptions to the AI landscape, predicting another “DeepSeek moment” within the next 18 months.