For example imagine asking "can you go into the other room, find object X, and bring it to me"
It seems to me the intelligence/reasoning ability is more than enough for something like that
And giving models images they seem to be able to recognize objects well to understand what they are looking at
But its not clear to me what is holding this product area back.
Is it a matter of the speed of the update loop or number of iterations it takes the agent being too slow to be practical?
Or does interpreting images/video cause the context/cost to blow up?
glenstein•1h ago
But in principle I think physical automation is the next big frontier and its probably coming within our lifetimes.