Ludo Robotics Inc., a subsidiary of major South Korean video game publisher Krafton Inc., today debuted Ludi 0.2, a robotics artificial intelligence system built to understand people and interact with them in the physical world.

Headquartered in Palo Alto, Ludo spun out of Krafton, a company that primarily publishes video games and is best known for the ultra-popular “PUBG: Battlegrounds.” Ludo also has a research and development lab in Seoul where it combines the efforts of researchers and engineers across robotics, machine learning and AI.

Krafton Chief Executive CH Kim also helms Ludo as CEO, with Krafton’s Chief AI Officer Kangwook Lee serving as its chief technology officer.

Why did a video game company form and back an AI robotics company? Krafton said it believes AI makes games more responsive and replayable. It also has strong roots in simulation, interaction and real-time systems, which have immediate applications in the bridge between the digital and physical, where the interaction between humans and robots lives.

“Robots are becoming increasingly capable of performing physical tasks, but understanding and responding to people remains a major challenge,” said Lee. “Ludi 0.2 represents another step toward our goal of bringing conversation, reasoning and physical action together so robots can better understand human intent, respond to feedback and adapt as situations change.”

The company introduced the first version of Ludi 0.1 in August 2026. It combines a vision-language model for reasoning and dialogue with specialized navigation and manipulation capabilities. Today’s next-generation model has advanced its capabilities, forming more natural human-robot interaction, including social intelligence, memory and reasoning, tool coordination, gesture interpretation, and physical movement.

The company said it’s working toward a final version of Ludi that will integrate perception, dialogue and physical action.

The objective is to integrate communication, reasoning and physical action into one “brain.” The model would watch what’s happening, listen to commands and speech, interpret those commands alongside human gestures and physical movements and reason about the environment. This would happen in real time, coordinating similarly to how another human decides on actions based on another person’s actions in their workspace.

This is the “Holy Grail” of physical AI, where intelligent robots can perform actions not just in industrial environments and warehouses, but also retail and domestic spaces such as restaurants, living rooms and kitchens.

Krafton said much of its work includes proprietary foundation models and co-playable characters developed in collaboration with Nvidia Corp., AI teammates designed to understand players in real time. Ludo builds on this foundation to develop the same kind of intelligence into real-world robots.

Startups are bringing the physical world to robots so they can see objects and grab and move them with increasing capability. Interpreting human intent and responding to social cues from people around the robot, however, remains elusive. Simulation makes it easy for physical AI to prepare robots to work with objects and around humans, but it doesn’t always prepare them to model human behavior.

Photo: Ludo Robotics