Universal Robots AI Trainer: Revolutionizing Robot Learning with AI & Cobots

Universal Robots (UR), a pioneer in collaborative robotics and a division of Teradyne Robotics, unveiled the UR AI Trainer at GTC 2026, a new system designed to accelerate the development of AI-powered robots. Developed in collaboration with Scale AI, the AI Trainer aims to bridge the gap between laboratory-based AI research and real-world factory deployment.

The system addresses a key challenge in robotics: the difficulty of translating AI models trained in controlled environments to the complexities of industrial settings. According to Anders Beck, VP of AI Robotics Products at Universal Robots, customers are increasingly seeking solutions to collect high-fidelity, synchronized robot and vision data for training AI models on the same robots they intend to deploy. “Our AI Trainer is the industry’s first direct lab-to-factory solution for AI model training,” Beck stated.

Traditional AI robot training often suffers from fragmented hardware and low-quality data collection. Much of the existing training data is gathered using research platforms that don’t accurately reflect the conditions found in production environments. Many systems rely solely on visual data, making it difficult to train robots for tasks requiring physical interaction and force feedback.

The UR AI Trainer utilizes a “leader-follower” setup. A human operator physically guides a “leader” robot through a task, while a synchronized “follower” robot mirrors the movements in real-time. Throughout the demonstration, the system simultaneously captures motion trajectories, force feedback data, and visual information, creating multimodal datasets suitable for training Vision-Language-Action (VLA) models.

A key differentiator of the UR AI Trainer is its ability to capture data directly on production-grade UR cobots. This allows developers to train models on the same hardware they will ultimately be deployed on, reducing the risk of performance degradation when transitioning from the lab to the factory floor. Universal Robots has deployed over 100,000 robots globally, providing a robust platform for long-running experiments and repeatable results.

The AI Trainer integrates with Scale AI’s software to facilitate the capture and management of this high-fidelity data. Deployed on Universal Robots’ AI Accelerator toolkit, the system creates a continuous data feedback loop, enabling ongoing optimization of AI-powered physical systems.

At GTC 2026, Universal Robots likewise showcased a robotics foundation model developed by Generalist AI, a strategic partner. The demonstration featured two UR robots performing a complex smartphone packaging task, a feat previously unattainable without recent advancements in physical AI.