Knowledge Production Based on Large Models and the Dialectic of Enlightenment
In the use of large models,there exists some degree of illusion that large models know everything and can do everything.It is precisely because of the existence of such super powers that large models may turn from rationality of enlightenment into new myths of enlightenment.The significance of enlightenment lies in maintaining a kind of openness,while the myth of enlightenment signifies the disappearance of this openness.The development of large models will undoubtedly intensify the mathematization of the world.Against this backdrop,credentials become the basic currency of the future digital world.Excessive mathematization also brews new dangers of instrumental rationality.As a new technology,large models fully demonstrate the relationship between knowledge and power.Large models possess tremendous integrative power and are realizing a new grand unification of knowledge.These diffused super powers will enter every field of knowledge production,with no individual able to escape the capture of these superpowers.Large models will further intensify the industrialization of knowledge,leading to more severe ideological issues.On the one hand,large models themselves have their own ideology.On the other hand,the deeper ideology of large models will be implicitly contained in their either free or consumption models,while the imagined knowledge community will eventually serve the hegemony of knowledge.Facing the risks of the industrialization of knowledge,"handcrafted oases"of knowledge production should be established,using absurdity,metaphor and rhetoric to confront rationality,allowing humans to retain the ability to create knowledge through embodied experiences,and avoiding the myths of sameness and absolute correctness.
large modelsartificial intelligenceChatGPTdialectic of enlightenmentinstrumental rationality
万方数据
维普
