Ask a robot to pick up an egg or a bowling ball, and it may not know how to adjust its grip to account for the different weights. A new smart glove allows a neural network to learn the shape and weight of an object just with sensors embedded into a glove, which could be put to use with robots in factories or homes, and can even teach us about our own human grip. "There's been a lot of hope that we'll be able to understand the human grasp someday and this will unlock our potential to create this dexterity in robots," says Subramanian Sundaram at the Massachusetts Institute of Technology. "Right now, it's hard to imagine a robot loading a dishwasher, but this kind of technology could transform things like that." He and his team attached force-sensitive film to the palms and fingers of a knitted glove and stitched a network of 64 conductive silver threads into it.