What exactly are you claiming here? That a handful of theorems about the limits of mathematics and provability somehow combine to show that the current LLM-based AI developments will inevitably live up to what is expected of them? And that this is obvious to a select few? That all seems unlikely, to say the least.