Good post. Tangentially related: I have thought recently that we really need to refresh the way that we explain these ideas to beginners in this new world of A

Good post. Tangentially related: I have thought recently that we really need to refresh the way that we explain these ideas to beginners in this new world of AI/ML doing everything. For instance, here is an excerpt from Scott Aaronson's excellent paper on the topic: "To illustrate, suppose we wanted to program a computer to create new Mozart-quality symphonies and Shakespeare-quality plays. If P = NP via a practical algorithm, then these feats would reduce to the seemingly easier problem of writing a computer program to recognize great works of art. And interestingly, P = NP might also help with the recognition problem: for example, by letting us train a neural network that reverse-engineered the expressed artistic preferences of hundreds of human experts. But how well that neural network would perform is an empirical question outside the scope of mathematics." This has been kind of the standard way to explain this stuff to beginners for the last 20 years or so. It is an excellent paper and I highly recommend reading. Aaronson also goes on to explain how these are imperfect metaphors, that they are just for building intuition, etc, and gives better technical details. But it's still a good intuition-builder regardless. Or at least it was. How do we explain all of this stuff now that we have StableDiffusion and ChatGPT? From here on, people are going to grow up in a world in which these things are commonplace, rather than it being questionable how physically possible they will ever be. The "empirical question" Aaronson talks about has largely been proven in the affirmative, so given that we are here, how do we explain this stuff now?