论人工智能的可控性:局限性分析

Q3 Computer Science

Journal of Cyber Security and Mobility Pub Date : 2022-05-25 DOI:10.13052/jcsm2245-1439.1132

Roman V. Yampolskiy

{"title":"论人工智能的可控性:局限性分析","authors":"Roman V. Yampolskiy","doi":"10.13052/jcsm2245-1439.1132","DOIUrl":null,"url":null,"abstract":"The invention of artificial general intelligence is predicted to cause a shift in the trajectory of human civilization. In order to reap the benefits and avoid the pitfalls of such a powerful technology it is important to be able to control it. However, the possibility of controlling artificial general intelligence and its more advanced version, superintelligence, has not been formally established. In this paper, we present arguments as well as supporting evidence from multiple domains indicating that advanced AI cannot be fully controlled. The consequences of uncontrollability of AI are discussed with respect to the future of humanity and research on AI, and AI safety and security.","PeriodicalId":37820,"journal":{"name":"Journal of Cyber Security and Mobility","volume":" ","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2022-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"6","resultStr":"{\"title\":\"On the Controllability of Artificial Intelligence: An Analysis of Limitations\",\"authors\":\"Roman V. Yampolskiy\",\"doi\":\"10.13052/jcsm2245-1439.1132\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The invention of artificial general intelligence is predicted to cause a shift in the trajectory of human civilization. In order to reap the benefits and avoid the pitfalls of such a powerful technology it is important to be able to control it. However, the possibility of controlling artificial general intelligence and its more advanced version, superintelligence, has not been formally established. In this paper, we present arguments as well as supporting evidence from multiple domains indicating that advanced AI cannot be fully controlled. The consequences of uncontrollability of AI are discussed with respect to the future of humanity and research on AI, and AI safety and security.\",\"PeriodicalId\":37820,\"journal\":{\"name\":\"Journal of Cyber Security and Mobility\",\"volume\":\" \",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2022-05-25\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"6\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Journal of Cyber Security and Mobility\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.13052/jcsm2245-1439.1132\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q3\",\"JCRName\":\"Computer Science\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Cyber Security and Mobility","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.13052/jcsm2245-1439.1132","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q3","JCRName":"Computer Science","Score":null,"Total":0}

引用次数: 6

摘要

通用人工智能的发明预计将导致人类文明轨迹的转变。为了获得这种强大技术的好处并避免其陷阱，能够控制它是很重要的。然而，控制通用人工智能及其更先进版本超级智能的可能性尚未正式确定。在本文中，我们提出了来自多个领域的论点和支持证据，表明高级人工智能无法完全控制。从人类的未来、人工智能的研究以及人工智能的安全和保障方面讨论了人工智能不可控的后果。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

On the Controllability of Artificial Intelligence: An Analysis of Limitations

The invention of artificial general intelligence is predicted to cause a shift in the trajectory of human civilization. In order to reap the benefits and avoid the pitfalls of such a powerful technology it is important to be able to control it. However, the possibility of controlling artificial general intelligence and its more advanced version, superintelligence, has not been formally established. In this paper, we present arguments as well as supporting evidence from multiple domains indicating that advanced AI cannot be fully controlled. The consequences of uncontrollability of AI are discussed with respect to the future of humanity and research on AI, and AI safety and security.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Journal of Cyber Security and Mobility Computer Science-Computer Networks and Communications

CiteScore

2.30

自引率

0.00%

发文量

期刊介绍： Journal of Cyber Security and Mobility is an international, open-access, peer reviewed journal publishing original research, review/survey, and tutorial papers on all cyber security fields including information, computer & network security, cryptography, digital forensics etc. but also interdisciplinary articles that cover privacy, ethical, legal, economical aspects of cyber security or emerging solutions drawn from other branches of science, for example, nature-inspired. The journal aims at becoming an international source of innovation and an essential reading for IT security professionals around the world by providing an in-depth and holistic view on all security spectrum and solutions ranging from practical to theoretical. Its goal is to bring together researchers and practitioners dealing with the diverse fields of cybersecurity and to cover topics that are equally valuable for professionals as well as for those new in the field from all sectors industry, commerce and academia. This journal covers diverse security issues in cyber space and solutions thereof. As cyber space has moved towards the wireless/mobile world, issues in wireless/mobile communications and those involving mobility aspects will also be published.