People have commented on this and said that the gist of the issue is mathematicians egos are bruised. That's true but I think it's defensible. Like we don't typically enjoy watching humans race against motorcycles in a 100m dash. But there's something interesting about watching humans do this in the Olympics when they race each other. Likewise, in math, people are solving problems considered hard by other humans and we acknowledged the art form.
But I think there's a practical side of it too. Finding what the limits of humans are is worthwhile too. Maybe the best analogy is like a group of people trying to find their way through a cave in the dark. One of them found a new passage and has a technique for doing it. So maybe we expanded what we thought humans were capable of. Could there be passages that humans can't get through? That might be useful to know. If we find proofs that humans just can't comprehend or have knowledge requirements that humans can't reasonably obtain that might be useful for us to know. A lot of math is not really applied to anything useful, it is just knowledge that is accumulated that other people thought was worthwhile perhaps because it was considered difficult for humans. So basically the journey to the proof is the value add and not the proof itself perhaps and AI might be erasing the journey. I have previously described it as a helicopter ride to the top of Everest but the point of climbing Everest is the difficult part. That's why people do it, the challenge.
But I think there's a practical side of it too. Finding what the limits of humans are is worthwhile too. Maybe the best analogy is like a group of people trying to find their way through a cave in the dark. One of them found a new passage and has a technique for doing it. So maybe we expanded what we thought humans were capable of. Could there be passages that humans can't get through? That might be useful to know. If we find proofs that humans just can't comprehend or have knowledge requirements that humans can't reasonably obtain that might be useful for us to know. A lot of math is not really applied to anything useful, it is just knowledge that is accumulated that other people thought was worthwhile perhaps because it was considered difficult for humans. So basically the journey to the proof is the value add and not the proof itself perhaps and AI might be erasing the journey. I have previously described it as a helicopter ride to the top of Everest but the point of climbing Everest is the difficult part. That's why people do it, the challenge.