The South Korean government plans to distribute 200 humanoid robots starting next year, but concerns have been raised about the project's effectiveness. The domestic artificial intelligence (AI) software, which essentially serves as the brain of the robots, is still in development.
According to a report from the National Assembly Budget Office on November 3, the Ministry of Trade, Industry and Energy has allocated 30 billion won (approximately $22.5 million) for the distribution of humanoid robots to universities and government-funded research institutions. The total investment is expected to reach 126 billion won by 2030, with plans to distribute 840 humanoid robots.
This initiative aims to create initial demand for the domestic humanoid industry and encourage companies to invest in mass production. The government intends to supply domestic robots to universities and research institutions for research and educational purposes while also training related professionals.
The subsidy for each robot is set at 150 million won. This includes 130 million won for hardware platforms, sensors, and robotic hands, and 20 million won for software platform integration and hardware linkage costs.
However, a significant issue is the lack of a domestic Robot Foundation Model (RFM), which is crucial technology for humanoids. The RFM is an AI model that learns from language, visual, and behavioral data, enabling robots to perceive their environment and make autonomous decisions.
Currently, there is no RFM available in South Korea that can be applied for research and educational purposes in humanoids. The Ministry of Trade, Industry and Energy has noted that major domestic universities and research labs primarily use Chinese hardware and American or open-source software, with no practical examples of domestic humanoids in use.
The government is also pursuing the development of the RFM, planning to invest 9.449 billion won by the end of 2028. Even if robot distribution begins next year, initial research and education may be limited due to reliance on foreign technology.
Proactively distributing humanoid hardware could help accumulate research experience and establish a practical foundation for domestic RFM. However, there are concerns that prioritizing hardware distribution alone may not lead to the self-sufficiency of domestic software.
The Budget Office emphasized, "If the distribution of humanoid hardware is prioritized without a corresponding focus on domestic RFM development, there is a risk that self-sufficiency in domestic RFM may not be achieved. It is essential to closely examine the development and validation of domestic RFM and ensure that the distribution of domestic hardware and software is interconnected in the project."
* This article has been translated by AI.
Copyright ⓒ Aju Press All rights reserved.