#quote #q #цитата #ai #ии #lang en
Will people lose interest in results that don't have far reaching implications, or were just low hanging fruits?

I think what a lot of people don't see is that we weren't "interested" in these to begin with. Like, for the vast majority of advances, very very few people are affected and most results are so niche as to be essentially worth nothing.

But... They still were genuine problems whose solutions demonstrated a certain ability. Our "interest" in these problems isn't necessarily the answer, but that doesn't mean they are without value. It's valuable to have a way to tell who understands mathematics well enough to make contributions. It may also be valuable to see if the proof involves new techniques/methods/patterns/ideas, things that could be used in other contexts (even if the result itself does not). And it cannot be measured how valuable it is that we have problems new mathematicians can train on in a way that allows them to search the entirety of literature without ever being given the answer. This is the hardest skill*. And w e can't do that anymore. We threw all that away in exchange for a trivia generation machine.

*Anecdote time. As a college teacher, let me just tell you that the number of grad students is going to fall to near zero soon, because the number of students who have the willpower to not use AI and actually learn Calculus is basically zero. I am not exaggerating when I say that I failed 90% this last semester. If we don't change things soon, the world is going to face an expertise crisis. These machines, despite how useful they are and how knowledgeable they present, are not experts. And without experts to learn from, there are very real and hard limits to what the next batch of experts can do.

(c) sqrtsqr, reddit.com/r/math/comments/1ux…