We prefer to take a high compute ideology for general problem solving over embedding human knowledge /bespoke methods for AI models because of obvious limitations to how human brain works. So why struggle so hard to achieve a human like AGI, isn't it counterproductive? Can someone help me here :’)