SAFETY OF ARTIFICIAL SUPERINTELLIGENCE
DOI:
https://doi.org/10.17770/etr2019vol2.4042Keywords:
Artificial Superintelligence, Cyber Security, Singularity, Safety of Artificial IntelligenceAbstract
The paper analyses an important problem of cyber security from human safety perspective which is usually described as data and/or computer safety itself without mentioning the human. There are numerous scientific predictions of creation of artificial superintelligence, which could arise in the near future. That is why the strong necessity for protection of such a system from causing any farm arises. This paper reviews approaches and methods already presented for solving this problem in a single article, analyses its results and provides future research directions.Downloads
References
R. Banham, Cybersecurity: Protective Measures Treasuries Should Be Taking. Treasury & Risk. 2018 Special Report, pp. 2-7.
D.Beskow, K.Carley, Social Cybersecurity: An Emerging National Security Requirement. Military Review. April 2019, Vol. 99 Issue 2, pp. 117-127. 11p.
N. Bostrom, Global Catastrophic Risks. Oxford: Oxford University Press, 2007.
N. Bostrom. The ethics of artificial intelligence. Cambridge Handbook of Artificial Intelligence, 2011. [Online]. Available: https://nickbostrom.com/ethics/artificial-intelligence.pdf [Accessed: March. 03, 2019].
R. Deibert, Toward a Human-Centric Approach to Cybersecurity. Ethics & International Affairs. Winter 2018, Vol. 32 Issue 4, pp. 411-424.
D. Gunning, Explainable Artificial Intelligence, DARPA project, 2018. [Online]. Available: https://www.darpa.mil/program/explainable-artificial-intelligence [Accessed: March. 07, 2019].
S. Hawking, Science in the next millenium, 1998. [Online]. Available:https://www.learnoutloud.com/Catalog/Science/Physics/Science-in-the-Next-Millennium/45223 [Accessed: March. 03, 2019].
N.Hennig, Privacy and Security Online: Best Practices for Cybersecurity. Library Technology Reports. April 2018, Vol. 54 Issue 3, pp. 1-37.
A. Holzinger, “From Machine Learning to Explainable AI.” World Symposium on Digital Intelligence for Systems and Machines August 2018. [Online]. Available: https://www.researchgate.net/publication/328309811_From_Machine_Learning_to_Explainable_AI [Accessed: Feb. 21, 2019].
M. Kiss, C. Muha, the cybersecurity capability aspects of smart government and industry 4.0 programmes. Interdisciplinary Description of Complex Systems. 2018, Vol. 16 Issue 3-A, pp. 313-319.
R. Kurzweil, The Singularity Is Near: When Humans Transcend Biology. New York, NY: Viking, 2006.
B. Lampson, A Note on the Confinement Problem, 1973. [Online]. Available: https://www.cs.utexas.edu/~shmat/courses/cs380s_fall09/lampson73.pdf [Accessed: March. 19, 2019]
Open AI project. [Online]. Available: https://openai.com/ [Accessed: March. 11, 2019].
N. Sales, Privatizing Cybersecurity. UCLA Law Review. April 2018, Vol. 65 Issue 3, pp. 620-688. 69p.
W. Samek, T. Wegang, K. Muller. Explainable artificial intelligence: understanding, Visualizing and interpreting deep learning models, Aug. 28, 2017. [Online]. Available: https://arxiv.org/abs/1708.08296 [Accessed: March. 19, 2019].
M.Scala, A. Reilly, Risk and the Five Hard Problems of Cybersecurity. Risk Analysis: An Official Publication Of The Society For Risk Analysis, March 2019, pp. 32-37.
A.T.Sherman, D. DeLatte, M.Neary etc., Cybersecurity: Exploring core concepts through six scenarios. Cryptologia. July 2018, Vol. 42 Issue 4, p337-377. 41p.
R. Yampolskiy, "Leakproofing the Singularity Artificial Intelligence Confinement Problem," Journal of Consciousness Studies, vol. 19, pp. 1-2, 2012.
R. Yampolskiy, Artificial Superintelligence: a Futuristic Approach. New York: Chapman and Hall/CRC, 2015.
E. Yudkowsky, The AI-Box experiment. [Online]. Available: http://yudkowsky.net/singularity/aibox/ [Accessed: March. 19, 2019].
