XDOF Raises $70 Million to Build Robotics AI Data Pipelines
Updated
Updated · TechCrunch · Jun 17
XDOF Raises $70 Million to Build Robotics AI Data Pipelines
1 articles · Updated · TechCrunch · Jun 17
Summary
$70 million in backing has brought XDOF out of stealth as a new infrastructure startup focused on collecting, cleaning and annotating training data for robotics and physical AI.
20 customers — including unnamed frontier AI labs — are already using its tools because robot models lack the large-scale, high-fidelity physical interaction data that language models got from public text.
130,000 manipulation trajectories anchor XDOF's first major dataset release, built with UC Berkeley and paired with 300 simulation hours and 100 evaluation hours for academic pre-training.
60 employees are building a labor-heavy pipeline spanning teleoperated robots and wearable sensors, betting major labs will outsource the warehouses, robot fleets and operator training needed to scale data production.
OpenAI's recent robotics relaunch underscores the broader race XDOF is targeting: turning scarce real-world robot experience into a repeatable data feedback loop for physical AI.
As XDOF builds a global workforce, what ethical standards will protect its human data collectors?
How will human-collected data effectively train robots with vastly different physical bodies and sensors?
Can a third-party data provider truly rival the proprietary datasets of vertically integrated robotics giants?
XDOF Secures $70 Million to Lead the Next Wave of Robotics AI with Scalable, High-Fidelity Data
Overview
XDOF has secured $70 million in funding to address the critical data bottleneck in robotics AI. As the robotics and physical AI sector rapidly expands, there is a growing need for high-quality, real-world data to develop advanced autonomous systems. XDOF’s unique, labor-intensive approach focuses on hiring and training global teams of teleoperators and egocentric data operators, directly tackling the significant gap in comprehensive data production. This investment highlights the industry’s recognition of the importance of specialized data infrastructure, positioning XDOF to play a key role in revolutionizing data for intelligent machines.