multi-modal models trained on videos and text already learn through 3 senses. google and others already connected these to arms (with very limited tactile infor

multi-modal models trained on videos and text already learn through 3 senses. google and others already connected these to arms (with very limited tactile information though). we wouldn't deny intelligence to people with disabilities. a model which was force-fed more than any human can ever experience is an alien intelligence, but the model it creates internally seems sufficient in many ways. if you accept the ability to do 'cold reasoning' as intelligence I'd say we are pretty close.