x
Are We Testing the Wrong Dimensions of AI Intelligence? — LessWrong