@simon @jimgar not sure I follow. Are you saying that crawler data, which includes copyrighted material shouldn’t be used by commercial companies and LLMs are inherently flawed because of that? If so, I’m not saying you’re wrong, just trying to understand.