论人工智能的自主性

The Autonomy of Artificial Intelligence

自主性由自我意识、自由意志和道德责任三个基本要件构成。根据自我意识概念,人工智能可以具有“形式自我意识”,不具有“实质自我意识”。根据自由意志概念,人工智能可以具有弱自由意志,但它只是决定论世界中的自由意志者,不具有同人类一样的能动性自由意志。只有能动性自由意志才能生发出创造性。因此,按以上论述,人工智能没有创造性。人工智能可以具有道德能力,但道德能力只是道德责任的必要而非充分的条件。承担道德责任必须具备理解自身的道德行为与道德准则之间的关系的能力,但人工智能不具备这种能力。因此,人工智能不能独立承担道德责任。根据构成自主性的三个基本要件衡量,人工智能只具有弱自主性。思考人工智能的自主性问题,既具有理论价值,也具有现实意义。

Autonomy is composed of three basic elements:self-awareness,free will,and moral responsibility.According to the concept of self-awareness,artificial intelligence can have“formal self-awareness”but not“substantial self-awareness”.According to the concept of free will,artificial intelligence can have a weak free will,but it is only a free will in a deterministic world and does not have the same active free will as human beings.Only active free will can produce creativity.Therefore,artificial intelligence is not creative.Artificial intelligence can have moral ability,which is only a necessary but not a sufficient condition for moral responsibility.To assume moral responsibility,there must be the ability to fully understand the relationship between one s own moral behavior and morality.However,artificial intelligence does not have this ability.Therefore,artificial intelligence cannot independently assume moral responsibility.According to the three basic elements of autonomy,artificial intelligence has only a weak autonomy.Reflecting on the autonomy of artificial intelligence has both theoretical value and practical significance.