I asked a mathematician the same question about whether we should allow AI systems to access all research papers and ideas, potentially putting their future car

I asked a mathematician the same question about whether we should allow AI systems to access all research papers and ideas, potentially putting their future careers at risk. He was very pessimistic about how human knowledge will compete with AI-generated proofs flooding the market. Especially in mathematics, many specialized areas have fewer than 100 people worldwide who are capable of determining whether a result is correct or not. The mathematics community will certainly be willing to use AI to assist with their research, but they may strongly object to a flood of AI-generated mathematics papers produced by others.