AIVEX Showcases “Physical AI Model and Data Closed-Loop Platform” at Autonomous Manufacturing AI World Show (AMWS 2026)

April 27, 2026, Seoul – AIVEX, an industrial Physical AI company, unveiled its “Physical AI Model and Data Closed-Loop Platform” during the keynote session at the Autonomous Manufacturing AI World Show (AMWS 2026) held at COEX on April 27.

AIVEX CEO Minsoo Sung emphasized the transition toward “action-oriented AI” that goes beyond decision-making to physical execution through a closed-loop autonomous manufacturing cycle integrating data, AI, and robotics.

Under the theme “Beyond Decision-Making to Action: The Direction of Autonomous Manufacturing Innovation Enabled by Physical AI,” CEO Sung highlighted the limitations of conventional manufacturing AI, which has primarily focused on perception and analysis, and explained the need for a Physical AI architecture that extends into real-world execution.

In particular, he identified key challenges preventing rapid AI transformation in manufacturing, including:
△ complex and dynamic industrial environments
△ difficulties in acquiring and utilizing manufacturing data
△ limitations in performance validation and operational reliability

To address these challenges, AIVEX introduced the concept of “Manufacturing Physical AI” as a new approach for industrial AI adoption.

AIVEX presented an end-to-end autonomous manufacturing cycle integrating AI vision, data platforms, and robot control technologies. The cycle consists of four core product platforms:
△ AIVision (AI Vision Inspection Platform)
△ AIVData (Data Platform)
△ AIVOps (MLOps Platform)
△ AIVot (AI Robotics Automation Platform)

The key differentiator of this architecture is its ability to connect the entire workflow—from data collection and model training to decision-making and physical execution—into a single continuous loop.

[Figure 1] AIVEX explained that it has built a unique AI vision ecosystem where data continuously circulates and improves, from data acquisition through AIVision, data refinement through AIVData, model training through AIVOps, and deployment back into the field through AIVision and AIVot.

Through this structure, AIVEX enables a data flywheel-based closed-loop system, where AI continuously learns from manufacturing data generated on-site and automatically incorporates improved intelligence into robot control and operations.

[Figure 2] AIVEX’s four core product platforms: AIVision, AIVot, AIVOps, and AIVData.

CEO Sung said, “AI is now moving beyond the stage of simply making decisions and entering an era where it can autonomously execute actions. Through a Physical AI-based structure that enables autonomous operation of real manufacturing processes, we aim to establish a new standard for manufacturing competitiveness.”

He added, “Manufacturing is the key competitive arena where AI technology directly translates into real productivity. We will continue advancing autonomous manufacturing through AI systems that are both performance-validated and reliably operational in industrial environments.”

During the afternoon track session, AIVEX will further introduce AI-based inspection automation, robotics application cases, and practical implementation approaches in real manufacturing environments.

Meanwhile, AIVEX changed its company name from AiV to AIVEX in January this year. The rebranding reflects its “Expansion” strategy, expanding beyond AI vision inspection into MLOps platforms and robotics while building a Physical AI ecosystem integrating data, AI, and robotics.