> I wonder if the money open ai and anthropic have spent just on solving these problems was spent on human mathematicians and resources would we have gotten these solved? Sooner?
Would the problems have been solved? Yes
Sooner? Probably not.
No government or private sector entity would have one-shotted the financing since it would have been part of their 'fundamental research' budget. Basic research is risky, is not guaranteed to provide any useful results and as a result tends to be constrained. Funding for basic research is money that you are lighting on fire.
Also for a given dollar spend, you cannot easily parallelize the efforts of top human mathematicians in the same way that you can with agents. Humans get tired, pass away, will withhold information etc.