%0 Journal Article %T 人工智能发展的宪法限度
Constitutional Limits of AI Development %A 吴璇 %J Open Journal of Legal Science %P 7132-7141 %@ 2329-7379 %D 2024 %I Hans Publishing %R 10.12677/ojls.2024.12121013 %X 人工智能发展应当给予一定限制，无序的发展可能会使人工智能走向消失的命途。人工智能研发者有失公正的输入数据行为系“算法歧视”问题产生的根本原因，衍生而来的“价格歧视”“种族歧视”“性别歧视”问题成为实现全人类平等目标的拦路虎；无序的人工智能发展对自由、安全等基本价值理念造成巨大冲击，宪法基本价值根基有所动摇；侵权认定难现象时有发生，人工智能成为侵权主体逃避法律责任的“避风港”。由于人工智能发展历史与宪法发展历史存在基本对应关系，宪法系当下解决人工智能失控所造成的法律问题的最佳手段。因此，应当将保障人的尊严这一宪法共识贯穿人工智能发展始终，使人工智能发展符合宪法发展目的——保障人类的主体性地位，在人工智能研发过程中坚持技术开发类型化的宪法基本原则，最终形成和平有序的宪法秩序来确保人工智能的有序发展。
The development of artificial intelligence should be given to certain restrictions, and the disorderly development may make artificial intelligence disappear. The unfair input data behavior of AI developers is the root cause of “algorithm discrimination”, and the derived “price discrimination”, “racial discrimination” and “gender discrimination” become the obstacle to the realization of equality; the disorderly development of AI has a huge impact on the basic values such as freedom and safety, and the foundation of the Constitution is shaken, and artificial intelligence becomes the “safe haven” for the infringing subject to evade legal responsibility. Due to the basic correspondence between the development history of artificial intelligence and the development history of the Constitution, the Constitution is the best means to solve the legal problems caused by the loss of control of artificial intelligence. Therefore, should safeguard the dignity of the constitutional consensus throughout the artificial intelligence development, artificial intelligence development in the constitutional purpose—safeguard the status of human subjectivity, in the process of artificial intelligence development type of basic principles of the constitution, eventually form a peaceful and orderly constitutional order to ensure the orderly development of artificial intelligence. %K 人工智能， %K 算法歧视， %K 人的尊严， %K 宪法共识
Artificial Intelligence %K Algorithmic Discrimination %K Human Dignity %K Constitutional Consensus %U http://www.hanspub.org/journal/PaperInformation.aspx?PaperID=103292