Abstract
The complex socio-technological debate underlying safety-critical and ethically relevant issues pertaining to AI development and deployment extends across heterogeneous research subfields and involves in part conflicting positions. In this context, it seems expedient to generate a minimalistic joint transdisciplinary basis disambiguating the references to specific subtypes of AI properties and risks for an error-correction in the transmission of ideas. In this paper, we introduce a high-level transdisciplinary system clustering of ethical distinction between antithetical clusters of Type I and Type II systems which extends a cybersecurity-oriented AI safety taxonomy with considerations from psychology. Moreover, we review relevant Type I AI risks, reflect upon possible epistemological origins of hypothetical Type II AI from a cognitive sciences perspective and discuss the related human moral perception. Strikingly, our nuanced transdisciplinary analysis yields the figurative formulation of the so-called AI safety paradox identifying AI control and value alignment as conjugate requirements in AI safety. Against this backdrop, we craft versatile multidisciplinary recommendations with ethical dimensions tailored to Type II AI safety. Overall, we suggest proactive and importantly corrective instead of prohibitive methods as common basis for both Type I and Type II AI safety.
S. Ziesche—Independent Researcher.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
AI risks of Type Ib have already been recognized in the AI field. However, risk Ib is still understudied for intelligent systems (often referred to as “autonomous” systems) deployed in real-world environments offering a wider attack surface.
- 2.
It is not contested that inductive inferences are logically invalid as shown by Popper. However, he also stated that “I hold that neither animals nor men use any procedure like induction, or any argument based on repetition of instances. The belief that we use induction is simply a mistake” [27] and that “induction simply does not exist” [27] (see [25] for an in-depth analysis of potential hereto related semantic misunderstandings). Arguments based on repetition of instances are existing but logically unfounded human habits as assumed by Hume [25], however they additionally require a point of view recognizing repetitions as such in the first place.
References
Aliman, N.M., Kester, L.: Artificial creativity augmentation. In: Goertzel, B., Panov, A.I., Potapov, A., Yampolskiy, R. (eds.) AGI 2020. LNCS (LNAI), vol. 12177, pp. 23–33. Springer, Cham (2020)
Aliman, N.M., Kester, L., Werkhoven, P., Ziesche, S.: Sustainable AI safety? Delphi Interdisc. Rev. Emerg. Technol. 2(4), 226–233 (2020)
Atzil, S., Gao, W., Fradkin, I., Barrett, L.F.: Growing a social brain. Nat. Hum. Behav. 2(9), 624–636 (2018)
Barrett, L.F.: The theory of constructed emotion: an active inference account of interoception and categorization. Soc. Cogn. Affect. Neurosci. 12(1), 1–23 (2017)
Barrett, L.F., Simmons, W.K.: Interoceptive predictions in the brain. Nat. Rev. Neurosci. 16(7), 419 (2015)
Baum, S.D.: Reconciliation between factions focused on near-term and long-term artificial intelligence. AI Soc. 33(4), 565–572 (2017). https://doi.org/10.1007/s00146-017-0734-3
Benedek, M.: The neuroscience of creative idea generation. In: Kapoula, Z., Volle, E., Renoult, J., Andreatta, M. (eds.) Exploring Transdisciplinarity in Art and Sciences, pp. 31–48. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-76054-4_2
Bieger, J., Thórisson, K.R., Wang, P.: Safe baby AGI. In: Bieger, J., Goertzel, B., Potapov, A. (eds.) AGI 2015. LNCS (LNAI), vol. 9205, pp. 46–49. Springer, Cham (2015). https://doi.org/10.1007/978-3-319-21365-1_5
Bigman, Y.E., Waytz, A., Alterovitz, R., Gray, K.: Holding robots responsible: the elements of machine morality. Trends Cogn. Sci. 23(5), 365–368 (2019)
Bostrom, N.: The superintelligent will: motivation and instrumental rationality in advanced artificial agents. Mind. Mach. 22(2), 71–85 (2012). https://doi.org/10.1007/s11023-012-9281-3
Brockman, J.: Possible Minds: Twenty-Five Ways of Looking at AI. Penguin Press, London (2019)
Bruineberg, J., Kiverstein, J., Rietveld, E.: The anticipating brain is not a scientist: the free-energy principle from an ecological-enactive perspective. Synthese 195(6), 2417–2444 (2016). https://doi.org/10.1007/s11229-016-1239-1
Clark, A., Friston, K., Wilkinson, S.: Bayesing qualia: consciousness as inference, not raw datum. J. Conscious. Stud. 26(9–10), 19–33 (2019)
De Rooij, A., Valtulina, J.: The predictive creative mind: a first look at spontaneous predictions and evaluations during idea generation. Front. Psychol. 10, 2465 (2019)
Deutsch, D.: Creative blocks. https://aeon.co/essays/how-close-are-we-to-creating-artificial-intelligence. Accessed Nov 2019
Deutsch, D.: The Beginning of Infinity: Explanations that Transform the World. Penguin, New York (2011)
Deutsch, D.: Constructor theory. Synthese 190(18), 4331–4359 (2013). https://doi.org/10.1007/s11229-013-0279-z
Dietrich, A.: How Creativity Happens in the Brain. Springer, London (2015). https://doi.org/10.1057/9781137501806
Friston, K.: Am I self-conscious? (Or does self-organization entail self-consciousness?). Front. Psychol. 9, 579 (2018)
Friston, K.: A free energy principle for a particular physics. arXiv preprint arXiv:1906.10184 (2019)
Goertzel, B.: The real reasons we don’ t have AGI yet. https://www.kurzweilai.net/the-real-reasons-we-dont-have-agi-yet. Accessed 21 Nov 2019
Goertzel, B.: Infusing advanced AGIs with human-like value systems: two theses. J. Evol. Technol. 26(1), 50–72 (2016)
Gray, K., Schein, C., Ward, A.F.: The myth of harmless wrongs in moral cognition: automatic dyadic completion from sin to suffering. J. Exp. Psychol. Gen. 143(4), 1600 (2014)
Gray, K., Wegner, D.M.: Feeling robots and human zombies: mind perception and the uncanny valley. Cognition 125(1), 125–130 (2012)
Greenland, S.: Induction versus popper: substance versus semantics. Int. J. Epidemiol. 27(4), 543–548 (1998)
Parr, T., Da Costa, L., Friston, K.: Markov blankets, information geometry and stochastic thermodynamics. Philos. Trans. R. Soc. A 378(2164), 20190159 (2019)
Popper, K.: In: Schilpp, P.A. (ed.) The Philosophy of Karl Popper, vol. 2, p. 1015. Open Court Press, Chicago (1974)
Popper, K.R.: The Poverty of Historicism. Routledge & Kegan Paul, Abingdon (1966)
Russell, S.: How to Stop Superhuman A.I. Before It Stops Us. https://www.nytimes.com/2019/10/08/opinion/artificial-intelligence.html?module=inline. Accessed 21 Nov 2019
Schein, C., Gray, K.: The theory of dyadic morality: reinventing moral judgment by redefining harm. Pers. Soc. Psychol. Rev. 22(1), 32–70 (2018)
Schulkin, J., Sterling, P.: Allostasis: a brain-centered, predictive mode of physiological regulation. Trends Neurosci. 42(10), 740–752 (2019)
Thórisson, K.R., Bieger, J., Li, X., Wang, P.: Cumulative learning. In: Hammer, P., Agrawal, P., Goertzel, B., Iklé, M. (eds.) AGI 2019. LNCS (LNAI), vol. 11654, pp. 198–208. Springer, Cham (2019). https://doi.org/10.1007/978-3-030-27005-6_20
Wang, P.: Motivation management in AGI systems. In: Bach, J., Goertzel, B., Iklé, M. (eds.) AGI 2012. LNCS (LNAI), vol. 7716, pp. 352–361. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-35506-6_36
Wiese, W.: Perceptual presence in the Kuhnian-Popperian Bayesian brain: a commentary on Anil K. Johannes Gutenberg-Universität Mainz, Seth (2016)
Yampolskiy, R.V.: Taxonomy of pathways to dangerous artificial intelligence. In: Workshops at the Thirtieth AAAI Conference on Artificial Intelligence (2016)
Acknowledgement
Nadisha-Marie Aliman would like to thank David Deutsch for providing a concise feedback on AI safety and Joscha Bach for a relevant exchange on AI ethics.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Aliman, NM. et al. (2020). Error-Correction for AI Safety. In: Goertzel, B., Panov, A., Potapov, A., Yampolskiy, R. (eds) Artificial General Intelligence. AGI 2020. Lecture Notes in Computer Science(), vol 12177. Springer, Cham. https://doi.org/10.1007/978-3-030-52152-3_2
Download citation
DOI: https://doi.org/10.1007/978-3-030-52152-3_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-52151-6
Online ISBN: 978-3-030-52152-3
eBook Packages: Computer ScienceComputer Science (R0)