As private corporations launch countless satellites, the center of gravity in the satellite industry is shifting from building better satellites to how quickly the deluge of data can be analyzed and turned into information. As a result, artificial intelligence (AI) is taking on a greater role in finding changes and extracting meaningful information from vast satellite imagery. A recommendation emerged that, for Korea to secure "K-space AI sovereignty," it needs systems and security frameworks that allow the use of high-quality data.
Park Won-gyu, executive vice president of Satrec Initiative (executive vice president of SIA), said on Aug. 26 at SMARTCLOUD SHOW 2026, the country's largest tech conference held at the Westin Josun Hotel in Jung District, Seoul, "Satellite imagery is exploding to a level that is difficult for people to interpret one by one," adding, "Now it is becoming difficult to interpret imagery without AI."
Satrec Initiative is connecting satellite systems, ground systems, imagery services, and AI-based geospatial analysis by its subsidiary SIA. It aims to cover the full process of converting "sensing" that observes the ground with satellites into "intelligence" that can be used for real-world decision-making.
Park said, "In the past, satellites were close to strategic assets developed over a long period with large-scale public funding and used for defense or disaster response," adding, "As the new space era has opened in which private corporations invest their own capital to operate multiple satellites, automated analysis through AI has become essential."
Satellite imagery combined with AI is already being used across fields such as disaster response, maritime, and weather. Park introduced cases in which AI analyzed satellite imagery taken after a tornado passed through Illinois in the United States to read the extent of building damage, and in which AI was applied to satellite data over African waters to identify ships.
Recently, cloud is also coming into satellite operations. Park said, "To receive data sent by satellites, you typically need a large antenna and ground station facilities costing in the billions of won," adding, "Services have emerged where corporations select only the hours they need to use ground stations that providers such as Amazon Web Services (AWS) have installed around the world."
Park cited defense and security as areas where satellites and AI combined can create significant added value. He introduced a future military operations scenario in which AI intervenes from the satellite imaging stage to find targets such as aircraft and ships and analyze anomalies. Further, AI can learn military doctrine and available weapons to analyze which means of attack would be effective and assess battle damage after an attack.
However, Park emphasized that to translate AI's potential into real industrial competitiveness, data and security issues must be addressed together. To secure "K-space AI sovereignty" by combining Korea's own satellites and AI technologies, systems and security frameworks must be overhauled together so that high-quality data can be used for AI training.
Park said, "Private corporations also secure data from satellites, but many of the high-quality data that can improve AI performance are held by government agencies, the military, and intelligence agencies," adding, "Using such data can produce excellent AI systems, but there are still hurdles to overcome."
Park also cited as a task finding ways to increase data utilization in the defense sector without compromising security. He said, "Even if we try to apply open-source AI or build a cloud environment, there are significant constraints in handling sensitive military data on external systems," adding, "We need to establish a framework that allows domestic corporations and researchers to build and validate AI while maintaining security."