{
  "id": 19581812,
  "title": "LG CNS Seeks Physical AI Paths With Experts",
  "url": "https://urgent.news/2026/11/12/lg-cns-seeks-physical-ai-paths-with-experts",
  "topic": "ai",
  "section": "AI",
  "published": "2026-11-12T01:30:51.000Z",
  "source": {
    "name": "BusinessKorea",
    "slug": "businesskorea",
    "url": "https://www.businesskorea.co.kr/news/articleView.html?idxno=281091"
  },
  "original_language": "en",
  "account": "Around 100 experts in robotics and artificial intelligence convened at a recent gathering, featuring faculty members and doctoral students from prestigious institutions such as the University of Toronto, Sogang University, Chung-Ang University, and Korea University. Additionally, representatives from LG CNS and its employees participated in the event. The attendees delved into technological trends in areas where robotics and AI intersect, such as reinforcement learning, robot mobility, intelligent mechanisms, and machine learning. They discussed the current state of physical AI demonstrations, including Robot Foundation Models, data utilization, intelligent robot hardware, and simulations, and engaged in detailed discussions on the technical hurdles encountered in real-world industrial settings.\n\nLG CNS shared insights into specific avenues for industrial applications, drawing from its extensive experience in executing over 1,000 smart factory and smart logistics projects over the past two decades. The company highlighted its technological prowess by showcasing two platforms: PhysicalWorks Forge, a robot learning platform introduced this year, and PhysicalWorks Baton, an agentic AI-based robot operation and control platform.\n\nGlobal research teams presented their latest accomplishments, followed by presentations by Professor Chi-geun Lee from the University of Toronto. Professor Lee elaborated on the development process of robot foundation models and the core technologies for their practical implementation. He discussed methods for integrating and leveraging data from multiple robots, as well as flow matching and action chunking technologies that facilitate continuous robot movements. Professor Lee also pointed out the challenges of movement variability among devices and the need to address data scarcity as a future concern.\n\nThe forum's participants examined physical AI applications across diverse sectors like logistics and manufacturing. They exchanged in-depth opinions on the operational tasks necessary for the dependable integration of robots, as well as the skills that companies and universities must cultivate, and the evolving roles of humans in the era of robots. Jun-ho Lee, Executive Vice President and Head of the Smart Logistics & City Business Division at LG CNS, emphasized, \"This forum served as a platform to exchange LG CNS's business experience with domestic and international researchers and to explore avenues for industrial application.\" He further stated, \"We will continue to collaborate with academia to bolster our business capabilities, ensuring that research breakthroughs translate into tangible innovations on-site.\"",
  "summary": "About 100 robotics and AI experts gathered at this event, including faculty members and master's and doctoral students from the University of Toronto, Sogang University, Chung-Ang University, and Korea University, as well as LG CNS executives and employees. The attendees reviewed technological trend",
  "key_points": [
    "100 experts in robotics and AI gathered at a conference.",
    "LG CNS showcased PhysicalWorks Forge and Baton platforms.",
    "Discussion focused on industrial applications and challenges."
  ],
  "editors_take": "Collaborating with academic researchers allows LG CNS to bridge the gap between theoretical breakthroughs in robot foundation models and the practical technical hurdles found in industrial manufacturing and logistics settings.",
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}