While I agree these terms are abused, the author exaggerates in the opposite direction. Granted, AI understood as AGI is "BS that doesn't exist" (yet?). But DL

While I agree these terms are abused, the author exaggerates in the opposite direction. Granted, AI understood as AGI is "BS that doesn't exist" (yet?). But DL and ML are working quite well and delivering decent results. It's true they're overhyped by unscrupulous marketers and greedy CEOs trying to grab more money from VC, but thanks to these technologies you can have pretty accurate text recognition, for example (and, to a lesser degree, although still impressive, also speech recognition).