You think ONE language model is going to cause all of this? What if it's not even THAT much better than it's predecessor (I mean it could be, but it could also not be, time will tell)? The AI researcher community hasn't even reached a consensus that language models are the way forward, so this comment strikes me as a bit optimistic. But like I said, time will tell if you're right or not.