Universal Robots has introduced its groundbreaking UR AI Trainer at GTC 2026 in Silicon Valley, developed in collaboration with Scale AI. The new system signals a major shift in robotics—from rigid, pre-programmed automation to adaptive, AI-driven machines capable of learning from human behavior and real-world interactions.
The UR AI Trainer is designed to bridge the long-standing gap between AI development in laboratories and deployment in real industrial environments. By enabling robots to learn directly from human-guided demonstrations, the system creates a seamless “lab-to-factory” pipeline for training and deploying AI models at scale.
Anders Beck, VP of AI Robotics Products at Universal Robots, stated:
“Our customers, ranging from large enterprises to AI research labs, are no longer just asking for AI features. They need a way to collect high-fidelity, synchronized robot and vision data to train 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.”
At the core of the UR AI Trainer is an advanced leader-follower system, where human operators guide a “leader” robot through tasks while a synchronized “follower” robot mirrors the actions in real time. This process captures high-quality multimodal data—including motion, force, and visual inputs—creating structured datasets essential for training Vision-Language-Action (VLA) models in robotics.
One of the system’s key innovations is its ability to capture high-fidelity physical interaction data. Traditional AI training often relies heavily on visual inputs, limiting performance in complex, contact-rich tasks. The UR AI Trainer overcomes this by incorporating Direct Torque Control and force feedback, allowing developers to precisely train how robots interact with objects in real-world scenarios.
The platform runs on Universal Robots’ AI Accelerator and integrates Scale AI’s data infrastructure, forming a continuous feedback loop where robots can be trained, deployed, and improved using real production data. This creates what Scale AI describes as a “robotics data flywheel,” enabling faster iteration and more robust AI model development.
Ben Levin, General Manager of Physical AI at Scale AI, emphasized:
“Universal Robots is a leader in industrial robotics, and its global footprint offers the ideal foundation for data capture and AI deployment at scale. Together, we’ve created an integrated robotics data flywheel, allowing customers to train, deploy, and improve their AI models faster than ever before.”
At GTC 2026, attendees experienced the system firsthand by guiding UR3e “leader” robots with haptic feedback to control UR7e “follower” robots. The demonstration showcased a complex smartphone packaging task—highlighting how imitation learning and real-time data capture can enable robots to perform intricate, previously difficult operations.
In addition to physical demonstrations, Universal Robots also showcased simulated training environments built using NVIDIA Omniverse and Isaac Sim. These virtual setups allow developers to generate physics-accurate synthetic data and train robots in digital environments before real-world deployment, further accelerating development cycles.
Amit Goel, Head of Robotics and Edge AI Ecosystem at NVIDIA, noted:
“The shift toward Physical AI requires a fundamental move from rigid, pre-programmed automation to generalist robots that can perceive, reason, and learn through human-like interaction. Universal Robots is building the infrastructure needed to train the next generation of autonomous systems at scale.”
Complementing the AI Trainer, Generalist AI demonstrated its robotic foundation models using UR robots to autonomously execute a complex smartphone packaging task. This real-world demonstration highlighted how high-quality data and advanced AI models can deliver enhanced dexterity, coordination, and reliability in industrial settings.
Pete Florence, Co-founder and CEO of Generalist AI, added:
“This demonstration shows how physical commonsense can be translated into real-world capability, paving the way for deployment across industries at scale.”
With over 100,000 robots deployed globally, Universal Robots is positioning itself as a key enabler of the Physical AI revolution. The UR AI Trainer represents a significant step toward scalable, intelligent automation—where robots continuously learn, adapt, and improve through real-world experience.
