I understand the appeal of having a machine helping us with maths and expanding the frontier of knowledge. They can assist researchers and make them more produc

I understand the appeal of having a machine helping us with maths and expanding the frontier of knowledge. They can assist researchers and make them more productive. Just like they can make already programmers more productive. But maths are also fun and fulfilling activity. Very often, when we learn a math theory, it's because we want to understand and gain intuition on the concepts, or we want to solve a puzzle (for which we can already look up the solution). Maybe it's similar to chess. We didn't develop search engines to replace human players and make them play together, but they helped us become better chess players or understanding the game better. So the recent progress is impressive, but I still don't see how we'll use this tech practically and what impacts it can have and in which fields.