It proves the limit doesn't exist, but doesn't seem to show how to find one of these more optimized limits. Given that nobody has ever brute-forced a counter-example, it may be that calculating these solutions is P=NP hard which would mean sticking with the status quo unless we find a good algorithm to calculate it on quantum computers and can actually get quantum computers large enough to work on the problem.

It's interesting as a math problem and test of AI, but not much else IMO.