🏋️‍♂️Physical AI - Training challenges

Large Language Models (LLMs) were trained on the entire internet's text. Physical AI on the other hand does not have a "Physical Internet" to learn from.

3D Data Scarcity: There is a massive shortage of high-quality, labeled 3D video and haptic (touch) data compared to text or 2D images or videos.

Physical Common Sense: Humans have intuition, e.g. they know that a glass will break if dropped or that a heavy box requires more force to push. Teaching this "intuitive physics" to AI requires a different kind of Model which holds a mental map of cause and effect It is still work in progress.