EMERGENCY EPISODE: Ex-Google Officer Finally Speaks Out On The Dangers Of AI! - Mo Gawdat | E252 - YouTube – I’m listening to this podcast, an interview with a former Googler, and this struck me:
And very soon they’ll come up with things we’ve never seen before.
But how?! if the transformers and the trillions of pages of text ChatGPT was trained on just predicts the next token/word, what possible mechanism will get to that next level of “Oh, now it learns and creates new things on its own”? 🤔 I don’t get it 🤷♂️