AI gets a body, and capital changes its gaze

Lee Soo-hwa The author is a research professor of Big Data Innovation Convergence College at Seoul National University and the head of the AI Center at DLG Law Corporation. AI's next battleground is not the data center but the living room. The valuation of Boston Dynamics, a robotics company under Hyundai Motor Group, has reportedly climbed to around 30 trillion won ($19.6 billion). Meanwhile, LG Electronics recently launched Axium, its robotics actuator series, and announced plans to become a vertically integrated robotics company. "The appliance of the future is a humanoid," Seoul National University Prof. Jang Byung-tak said, capturing a shift that is reshaping investor sentiment. This is not just another technology stock rally. Investors in western Seoul's Yeouido and on Wall Street are placing a premium on the moment when AI, once confined to generating text on screens, enters the physical world through machines equipped with motors and metal frames. Behind this shift is a new understanding of intelligence. Intelligence is no longer viewed as the product of calculation alone. Humans do not record the world as cameras do; instead, they create meaning through interactions shaped by perception, experience and emotion — a process that cognitive science refers to as embodied cognition. Industrial robotic arms operate in highly controlled settings, where they repeat predefined tasks. Future humanoids, however, must function in far less predictable environments. A family living room, where children leave their toys scattered across the floor, presents challenges that do not exist on an assembly line. As a result, the humanoid race involves more than computing power. People constantly create unexpected situations. They can alter the location of furniture and other household items. They can dirty the floor. Even the seemingly simple act of washing dishes could require adaptation. The longstanding AI ambition of categorizing the world and governing it through fixed rules is insufficient for such conditions. Humanoids must learn by interacting with reality rather than relying solely on preprogrammed responses. Competition has already become global. China's Unitree is pursuing aggressive expansion with relatively affordable humanoid robots. Tesla and major U.S. technology companies have begun deploying robots in their own factories. Korean firms are responding with advanced manufacturing capabilities. The key question is not who possesses the most sophisticated algorithm but who can build machines capable of functioning in imperfect human environments. The standards used to assess corporate value are also changing. Investors increasingly reward companies that move beyond controlled settings and into everyday life. The premium goes to those capable of recovering after failure and adapting to unexpected variables. Machines are no longer digital ghosts confined to screens. They are entering daily life in physical form. The future will belong to machines and organizations that can withstand real-world shocks and continue learning from them. A machine that never falls can never learn how to walk. This article was originally written in Korean and translated by a bilingual reporter with the help of generative AI tools. It was then edited by a native English-speaking editor. All AI-assisted translations are reviewed and refined by our newsroom.