I was just thinking with them being trained on Text, Images, Audios, Videos ... shouldn't they be called LIMs - Large Information Models?