期刊首页 优先出版 当期阅读 过刊浏览 作者中心 关于期刊 English

《中国工程科学》 >> 2021年 第23卷 第3期 doi: 10.15302/J-SSCAE-2021.03.005

针对强人工智能安全风险的技术应对策略

北京大学计算机科学技术系,北京100871

资助项目 :中国工程院咨询项目“新一代人工智能安全与自主可控发展战略研究” (2019-ZD-01) 收稿日期: 2021-04-07 修回日期: 2021-04-25 发布日期: 2021-06-01

下一篇 上一篇

摘要

未来进入强人工智能(AGI)时代,人类可能面临重大安全风险。本文归纳了AGI 与传统人工智能的区别,从模型的不可解释性、算法及硬件的不可靠性、自主意识的不可控性三方面研判了AGI 安全风险的来源,从能力、动机、行为3 个维度提出了针对AGI 的安全风险评估体系。为应对安全风险,从理论及技术研究、应用两个层面分别探讨相应风险的防御策略:在理论技术研究阶段,完善理论基础验证,实现模型可解释性,严格限制AGI 底层价值取向,促进技术标准化;在应用阶段,预防人为造成的安全问题,对AGI 进行动机选择,为AGI 赋予人类价值观。此外,建议加强国际合作,培养强AI 研究人才,为迎接未知的强AI 时代做好充分准备。

参考文献

[ 1 ] 陈俊波, 高杨帆. 系统论视角下的人工智能与人类智能 [J]. 自然 辩证法研究, 2019, 35(9): 99–104. Chen J B, Gao Y F. Artificial intelligence and human intelligence from the perspective of system theory [J]. Studies in Dialectics of Nature, 2019, 35(9): 99–104.
Chen J B, Gao Y F. Artificial intelligence and human intelligence from the perspective of system theory [J]. Studies in Dialectics of Nature, 2019, 35(9): 99–104. Chinese. 链接1

[ 2 ] 黄铁军, 余肇飞, 刘怡俊. 类脑机的思想与体系结构综述 [J]. 计 算机研究与发展, 2019, 56(6): 1133–1148. Huang T J, Yu Z F, Liu Y J. Brain-like machine: Thought and architecture [J]. Journal of Computer Research and Development, 2019, 56(6): 1133–1148.
Huang T J, Yu Z F, Liu Y J. Brain-like machine: Thought and architecture [J]. Journal of Computer Research and Development, 2019, 56(6): 1133–1148. Chinese. 链接1

[ 3 ] 张钹. 走向真正的人工智能 [J]. 卫星与网络, 2018 (6): 24–27. Zhang B. Towards the real artificial intelligence [J]. Satellite & Network, 2018 (6): 24–27.
Zhang B. Towards the real artificial intelligence [J]. Satellite & Network, 2018 (6): 24–27. Chinese. 链接1

[ 4 ] 徐宗本 . AI与数学 “融通共进 ” 迈向自主智能时代 [EB/ OL]. (2020-06-08)[2021-02-15]. http://news.sciencenet.cn/ htmlnews/2020/6/441057.shtm. Xu Z B. AI and math go together towards the era of autonomous intelligence [EB/OL]. (2020-06-08)[2021-02-15]. http://news. sciencenet.cn/htmlnews/2020/6/441057.shtm.
Xu Z B. AI and math go together towards the era of autonomous intelligence [EB/OL]. (2020-06-08) [2021-02-15]. http://news. sciencenet.cn/htmlnews/2020/6/441057.shtm. Chinese. 链接1

[ 5 ] 周志华. 关于强人工智能 [J]. 中国计算机学会通讯, 2018, 14(1): 45–46. Zhou Z H. Views on artificial general intelligence [J]. Communication of the CCF, 2018, 14(1): 45–46.
Zhou Z H. Views on artificial general intelligence [J]. Communication of the CCF, 2018, 14(1): 45–46. Chinese.

[ 6 ] 黄铁军. 也谈强人工智能 [J]. 中国计算机学会通讯, 2018, 14(2): 47–48. Huang T J. Different views on artificial general intelligence [J]. Communication of the CCF, 2018, 14(2): 47–48.
Huang T J. Different views on artificial general intelligence [J]. Communication of the CCF, 2018, 14(2): 47–48. Chinese.

[ 7 ] Amodei D, Olah C, Steinhardt J, et al. Concrete problems in AI safety [EB/OL]. (2016-07-25)[2021-02-15]. https://arxiv.org/ abs/1606.06565. 链接1

[ 8 ] Congress of the United States. H.R.5356-National security commission artificial intelligence act of 2018 [EB/OL]. (2018-03-20) [2021-02-15]. https://www.congress.org/bill/115th-congress/housebill/5356. 链接1

[ 9 ] 中国信息通信研究院. 全球人工智能治理体系报告 [EB/ OL]. (2020-12-30)[2021-02-15]. https://pdf.dfcfw.com/pdf/H3_ AP202012301445361107_1.pdf?1609356816000.pdf. China Academy of Information and Communications Technology. Global AI governance report [EB/OL]. (2020-12-30)[2021-02- 15]. https://pdf.dfcfw.com/pdf/H3_AP202012301445361107_1. pdf?1609356816000.pdf.
China Academy of Information and Communications Technology. Global AI governance report [EB/OL]. (2020-12-30) [2021-02- 15]. https://pdf.dfcfw.com/pdf/H3_AP202012301445361107_1.pdf?1609356816000.pdf. Chinese. 链接1

[10] 金晶, 秦浩, 戴朝霞. 美国人工智能安全顶层战略及重点机构研 发现状 [J]. 网信军民融合, 2020 (5): 45–48. Jin J, Qin H, Dai Z X. Top-level strategy of artificial intelligence security and the research status of key institutions in the United States [J]. Civil-Military Integration on Cyberspace, 2020 (5): 45–48.
Jin J, Qin H, Dai Z X. Top-level strategy of artificial intelligence security and the research status of key institutions in the United States [J]. Civil-Military Integration on Cyberspace, 2020 (5): 45–48. Chinese. 链接1

[11] Whyte C. Deepfake news: AI-enabled disinformation as a multilevel public policy challenge [J]. Journal of Cyber Policy, 2020, 5(2): 1–19. 链接1

[12] Goodfellow I, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks [J]. Advances in Neural Information Processing Systems, 2014, 3(11): 2672–2680.

[13] Bau D, Zhu J Y, Wulff J, et al. Seeing what a GAN cannot generate [C]. Seoul: 2019 IEEE/CVF International Conference on Computer Vision, 2019.

[14] Huang T J. Imitating the brain with neurocomputer a “new” way towards artificial general intelligence [J]. International Journal of Automation and Computing, 2017, 14(5): 520–531. 链接1

[15] 曲晶, 张绿云. 国外火箭发射及故障情况统计分析 [J].中国航天, 2016 (2): 13–18. Qu J, Zhang L Y. Statistical analysis of foreign rocket launch and failure [J]. Aerospace China, 2016 (2): 13–18.
Qu J, Zhang L Y. Statistical analysis of foreign rocket launch and failure [J]. Aerospace China, 2016 (2): 13–18. Chinese. 链接1

[16] 邢会强. 证券期货市场高频交易的法律监管框架研究 [J]. 中国 法学, 2016 (5): 156–177. Xing H Q. Research on the legal regulatory framework of high frequency trading in securities and futures market [J]. China Legal Science, 2016 (5): 156–177.
Xing H Q. Research on the legal regulatory framework of high frequency trading in securities and futures market [J]. China Legal Science, 2016 (5): 156–177. Chinese. 链接1

[17] Tegmark M. Life 3.0: Being human in the age of artificial intelligence [M]. New York: Penguin Random House LLC, 2017.

[18] Bostrom N. Superintelligence: Paths, dangers, strategies [M]. Oxford: Oxford University Press, 2015.

[19] Vilalta R, Drissi Y. A perspective view and survey of meta-learning [J]. Artificial Intelligence Review, 2002, 18(2): 77–95. 链接1

[20] 李旭嵘, 纪守领, 吴春明, 等. 深度伪造与检测技术综述 [J]. 软件 学报, 2021, 32(2): 496–518. Li X R, Ji S L, Wu C M, et al. Survey on deepfakes and detection techniques [J]. Journal of Software, 2021, 32(2): 496–518.
Li X R, Ji S L, Wu C M, et al. Survey on deepfakes and detection techniques [J]. Journal of Software, 2021, 32(2): 496–518. 链接1

[21] Asimov I. I, robot [M]. Louisville: Spectra Press and Promotions, 2004.
Asimov I. I, robot [M]. Louisville: Spectra Press and Promotions, 2004. Chinese.

[22] 黄铁军. 人类能制造出“超级大脑”吗? [N]. 中华读书报, 2015- 01-07(5). Huang T J. Can human build “super brain”? [N]. China Reading Weekly, 2015-01-07(5).
Huang T J. Can human build “super brain”? [N]. China Reading Weekly, 2015-01-07(5). Chinese.

[23] 中华人民共和国科学技术部. 欧洲25国签署《人工智能合作宣 言》 [EB/OL]. (2018-07-18)[2021-02-15]. http://www.most.gov.cn/ gnwkjdt/201807/t20180718_140708.htm. Ministry of Science and Technology of the People’s Republic of China. 25 European countries sign the Declaration on Artificial Intelligence Cooperation [EB/OL]. (2018-07-18)[2021-02-15]. http://www.most.gov.cn/gnwkjdt/201807/t20180718_140708.htm.
Ministry of Science and Technology of the People’s Republic of China. 25 European countries sign the Declaration on Artificial Intelligence Cooperation [EB/OL]. (2018-07-18) [2021-02-15]. http://www.most.gov.cn/gnwkjdt/201807/t20180718_140708.htm. Chinese. 链接1

相关研究