2026-06-30T11:07:01
(BPT) –
By Jae-cheol Lyu, President & CEO of LG Electronics (LG)
NVIDIA, which first introduced the concept of Physical AI, is driving a shift in the industry by expanding AI from the digital world into the physical world. Physical AI refers to a technological paradigm that extends AI’s reasoning capabilities into the real world, enabling systems to perceive space, understand context, make independent judgments and take physical action. The core challenge of this is how to realize “autonomous agents” that can interact with our living environments and make a meaningful difference in our daily lives.
Against this backdrop, LG Electronics (LG) drew attention during NVIDIA founder and CEO Jensen Huang’s recent visit to South Korea for our differentiated competitiveness — connecting data, spaces, devices and industrial infrastructure to bring Physical AI to life in the real world.
LG Electronics and NVIDIA have already further defined the scope of collaboration and strategic partnership opportunities through follow-up discussions in the United States. We agreed to establish a close working framework to accelerate execution, identify mutually beneficial initiatives and jointly expand the Physical AI ecosystem. A key focus is collaboration around robot form factors that could become representative devices for Physical AI.
Beyond robotics, we are also discussing broader synergies, including an LG data factory leveraging NVIDIA’s platform, the advancement of LG’s AI data center cooling solutions and LG’s robot mass-production system. Through these expert-level conversations, we are accelerating our preparations for the future of the Physical AI ecosystem.
So why are global tech giants like NVIDIA turning to LG Group, with LG Electronics at its center, at this pivotal point in the evolution of AI?
Physical AI is a technology that connects AI to action in real-world spaces, and the ability to understand spaces, control devices and create tangible value is emerging as a new source of competitiveness. In this new opportunity space, LG Electronics is gaining recognition for its differentiated strengths connecting data, spaces, devices and industrial infrastructure to bring AI into the real world.
Real-World Data — The Fuel of Physical AI
Physical AI goes beyond the ability to understand language. It can recognize, reason and act within real-world environments. Reaching this new reality will require massive, high-quality data from the physical world.
At LG Electronics, we have collected an expansive repository of safely anonymized usage data from our diverse and constantly growing ThinQ™ portfolio, which is fed into a continuous learning structure that helps us gain insight into customer usage patterns. These insights are then reflected in LG products and solutions. In point of fact, we have accumulated more data than the Hubble Space Telescope.1 The data we collect can be deployed throughout the business value chain. It serves as a “human behavior signal,” revealing how people live and behave in real-world spaces.
These signals, alongside our world-class manufacturing capabilities, give us a distinct competitive edge when it comes to translating AI into real-world products and services.
Beyond this, we have access to detailed manufacturing data from our 31 production facilities, spanning 14 countries around the world. The volume of manufacturing and production data we have accumulated over the past decade alone reaches 770 terabytes, which is comparable to the storage capacity needed for approximately 197,000 high-definition films.2
In the era of Physical AI, this breadth and fidelity of data is a major advantage.
Spatial Expertise — The Map of Physical AI
Our deep insight into and understanding of diverse living spaces have long been key LG competitive strengths. To deliver Physical AI, the ability to understand how people live, move and interact with technology across various environments, and to adapt and orchestrate those spaces accordingly, is a crucial strategic advantage.
For decades, we have designed and operated in a wide range of spaces, including residential, commercial, industrial and mobility settings, while maintaining continuous customer touchpoints. Through this, we have developed a thorough understanding of people’s lifestyles, device usage patterns, movement flows within spaces and energy consumption behaviors. Building on our technological strengths in HVAC, mobility and robotics, we have secured the capabilities necessary to engineer and run real-world environments where AI solutions can function optimally and create meaningful value for customers.
Our rich knowledge of physical spaces isn’t defined only by matters of engineering; rather, it is equally informed by the human experience. A representative example is the Life Soft Research (LSR) Customer Research Institute, one of our core think tanks, that has spent more than 35 years studying customers around the world — including North America, Asia, the Middle East and Europe — examining their lifestyles, spatial behavior, consumer psychology and attitudes toward technology adoption to forecast the future lifestyles. Drawing on extensive expertise including psychology, anthropology, design, engineering and sociology, the institute has developed a nuanced comprehension of how customers live, move and interact with technology within different spaces. More recently, it has been utilizing AI and big data analysis to gain even more insight into behavioral context and customers’ hidden needs. These insights form the foundation of new product, service and customer experience innovation.
We believe that one of the true differentiators in the emerging Physical AI landscape is the ability to design the stage where AI can operate safely and efficiently. While Physical AI and humanoid robots may feel like science fiction right now, they will ultimately become critical components of our everyday environments.
Here, we hold a unique position in understanding how people interact with a wide range of spaces — including residential, commercial environments, industrial sites and mobility — and turning those insights into products and services. In the era of Physical AI, this experience is expected to become a distinctive strength in the global market.
Execution Capabilities to Make AI Real Technology Ecosystem
Ultimately, the purpose of Physical AI is to move and operate in the real world. This requires various means of physical execution, including sensors, motors, actuators and robotics. With the competitiveness of our state-of-the-art components, including the LG Actuator AXIUM™ and world-class manufacturing capabilities, we essentially command the entire value chain needed to bring Physical AI to life. In robotics in particular, we are laying the foundation for a next-generation Physical AI ecosystem by combining NVIDIA’s AI platform with our proprietary robotics technology and manufacturing know-how. In the AI infrastructure domain, we are supporting the development of more efficient and stable AI operating environments through technology collaboration in next-generation cooling solutions and AI data centers.
What further sets us apart in this key transitional phase is the ability to organically connect AI capabilities across the broader LG Group. Through collaboration with affiliates including LG AI Research, LG CNS, LG U+ and LG Energy Solution, we are integrating AI models, computing infrastructure, energy, data centers, software and operational capabilities into a cohesive whole. This enables us to deliver tailored end-to-end solutions for customers — from AI development and infrastructure build-out, to implementation in industrial settings.
Ultimately, what global big-tech firms like NVIDIA need is a partner capable of deploying, operating and scaling AI within physical environments. As a company that simultaneously possesses a wealth of real-world data, spatial expertise, execution capabilities and a “full stack” AI ecosystem, we are uniquely positioned to be the definitive partner in the Physical AI era.
As AI continues to evolve, global demand for the execution capabilities needed to realize — and scale — real-world AI will only grow. Together with leading companies like NVIDIA, we will connect AI to tangible customer value and industrial innovation, turning the possibilities and potential of Physical AI into reality.
1 Hubble by the Numbers – NASA Science: 477TB of data accumulated by the Hubble Telescope since 1990.
2 Assuming 4GB per film.
