According to TechCrunch, physical AI—robots learning to perceive and act in the real world—faces a uniquely unglamorous problem: training data. While language models trained on internet scraps, robots need real-world examples collected by humans. That's why AI labs are now outsourcing to specialized companies like XDOF to gather this data at scale. It's a reminder that even at the frontier of AI research, someone has to do the hands-on legwork.

Listen to this story

Hear this and more stories in a personalized audio briefing.

Open The Chonkerton