Larks comments on Comparative and absolute advantage in AI