Abstract
Some philosophers and machine learning experts have speculated that superintelligent Artificial Intelligences (AIs), if and when they arrive on the scene, will wrestle away power from humans, with potentially catastrophic consequences. Dan Hendrycks has recently buttressed such worries by arguing that AI systems will undergo evolution by natural selection, which will endow them with instinctive drives for self-preservation, dominance and resource accumulation that are typical of evolved creatures. In this paper, we argue that this argument is not compelling as it stands. Evolutionary processes, as we point out, can be more or less Darwinian along a number of dimensions. Making use of Peter Godfrey-Smith’s framework of Darwinian spaces, we argue that the more evolution is top-down, directed and driven by intelligent agency, the less paradigmatically Darwinian it becomes. We then apply the concept of “domestication” to AI evolution, which, although theoretically satisfying the minimal definition of natural selection, is channeled through the minds of fore-sighted and intelligent agents, based on selection criteria desirable to them (which could be traits like docility, obedience and non-aggression). In the presence of such intelligent planning, it is not clear that selection of AIs, even selection in a competitive and ruthless market environment, will end up favoring “selfish” traits. In the end, however, we do agree with Hendrycks’ conditionally: If superintelligent AIs end up “going feral” and competing in a truly Darwinian fashion, reproducing autonomously and without human supervision, this could pose a grave danger to human societies.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Explore related subjects
Discover the latest articles, news and stories from top researchers in related subjects.Data availability
Not applicable.
Notes
“Ilya: the AI scientist shaping the world” (The Guardian, 2023), conversations recorded between 2016 and 2019. bit.ly/46SIq83.
We are grateful to an anonymous reviewer for this observation.
This point about small genetic variations does not rule out relatively large saltations in phenotypes, which can sometimes be caused by a single point mutation (e.g. an extra limb caused by a mutation in a hox gene encoding positional information of limbs). Large genetic variations can also occasionally happen in a single generation, for instance through the duplication of a whole chromosome or large gene segments, but this is not typical.
References
Bongard, J. C. (2013). Evolutionary robotics. Communications of the ACM, 56(8), 74–83.
Bostrom, N. (2014). Superintelligence: Paths, dangers, strategies. Oxford University Press.
Boudry, M. (2018). Replicate after reading: On the extraction and evocation of cultural information. Biology & Philosophy, 33(3), 27.
Boudry, M., & Hofhuis, S. (2018). Parasites of the mind. Why cultural theorists need the Meme’s Eye View. Cognitive Systems Research, 52, 155–167. http://philsci-archive.pitt.edu/14691/
Bricken, T., Templeton, A., Batson, J., Chen, B., Jermyn, A., Conerly, T., Turner, N. L., Anil, C., Denison, C., Askell, A., Lasenby, R., Wu, Y., Kravec, S., Schiefer, N., Maxwell, T., Joseph, N., Tamkin, A., Nguyen, K., McLean, B., Burke, J. E., Hume, T., Carter, S., Henighan, T., & Olah, C. (2023). Towards monosemanticity: decomposing language models with dictionary learning, Anthropic research paper, released 4 October 2023, https://transformer-circuits.pub/2023/monosemantic-features/index.html
Butler, S. (1863). Darwin among the machines. The Press, June, 13(1863), 205.
Carlsmith, J. (2023). Scheming AIs: Will AIs fake alignment during training to get power? arXiv preprint arXiv:2303.08379.
Cotra, A. (2022). Without specific countermeasures, the easiest path to transformative AI likely leads to AI takeover. URL https://www.alignmentforum.org/posts/pRkFkzwKZ2zfa3R6H/without-specific-countermeasures-the-easiest-path-to
Darwin, C. (1859). On the origin of species. https://www.gutenberg.org/files/2009/2009-h/2009-h.htm
Darwin, C. (1871). The descent of man, and selection in relation to sex. John Murray.
Dawkins, R. (1976). The selfish gene. Oxford University Press.
Dawkins, R. (1983). Universal darwinism. In D. S. Bendall (Ed.), Evolution from molecules to man (pp. 403–425). Cambridge University Press.
Dawkins, R. (1986). The blind watchmaker. Longman Scientific & Technical.
Dennett, D. C. (1987). The intentional stance. MIT Press.
Dennett, D. C. (1995). Darwin’s dangerous idea: evolution and the meanings of life. Simon & Schuster.
Dennett, D. C. (2007). Breaking the spell: Religion as a natural phenomenon. Penguin UK. https://books.google.nl/books?id=e2eVSvJieC0C
Dennett, D. C. (2017). From Bacteria to Bach and back: The evolution of minds. Penguin Books. https://books.google.be/books?id=iHtEvgAACAAJ
Dennett, D. C. (2023). ‘’The Problem With Counterfeit People’. The Atlantic, May 16, 2023. https://www.theatlantic.com/technology/archive/2023/05/problem-counterfeit-people/674075/
Domingos, P. (2015). The master algorithm: How the quest for the ultimate learning machine will remake our world. Basic Books.
Driscoll, C. A., Macdonald, D. W., & O’Brien, S. J. (2009). From wild animals to domestic pets, an evolutionary view of domestication. Proceedings of the National Academy of Sciences, 106(supplement_1), 9971–9978. https://doi.org/10.1073/pnas.0901586106
Floreano, D., & Mattiussi, C. (2008). Bio-inspired artificial intelligence: Theories, methods, and technologies. MIT Press.
Friederich, S. (2023). Symbiosis, not alignment, as the goal for liberal democracies in the transition to artificial general intelligence. AI and Ethics, 4, 315–324.
Friederich, S., & Boudry, M. (2022). Ethics of nuclear energy in times of climate change: Escaping the collective action problem. Philosophy & Technology, 35(2), 30.
Fuhrmann, M. (2009). Spreading temptation: Proliferation and peaceful nuclear cooperation agreements. International Security, 34(1), 7–41.
Gibbons, R. D. (2020). Supply to deny: The benefits of nuclear assistance for nuclear nonproliferation. Journal of Global Security Studies, 5(2), 282–298.
Gibbons, D. R. (2022). The hegemon’s toolkit: US leadership and the politics of the nuclear nonproliferation regime. Cornell University Press.
Godfrey-Smith, P. (2009). Darwinian populations and natural selection. Oxford University Press.
Hendrycks, D. (2023). Natural selection favors AIs over humans. arXiv preprint arXiv:2303.16200.
Henrich, J. (2015). The secret of our success: How culture is driving human evolution, domesticating our species, and making us smarter. Princeton University Press.
Hodgson, G. M., & Knudsen, T. (2012). Darwin’s conjecture: the search for general principles of social and economic evolution. The University of Chicago Press.
Horner, A., & Goldberg, D. E. (1991). Genetic algorithms and computer-assisted music composition (Vol. 51). Michigan Publishing, University of Michigan Library.
Kunkel, T. A., & Bebenek, K. (2000). DNA replication fidelity. Annual Review of Biochemistry, 69(1), 497–529.
Lang, P. A. (2017). Nuclear power learning and deployment rates; disruption and global benefits forgone. Energies, 10, 2169.
Lewens, T. (2015). Cultural evolution: Conceptual challenges. OUP Oxford.
Lewontin, R. C. (1970). The units of selection. Annual Review of Ecology and Systematics, 1(1), 1–18.
Lewontin, R. C. (1985). Adaptation. In R. Levins, & R. C. Lewontin (Eds.), The Dialectical biologist (pp. 65–84). Harvard University Press.
Lu, M. (2021). This is how car safety improved over the last 60 years, World Economic Forum, https://www.weforum.org/agenda/2021/12/how-safety-improved-over-60-years/, accessed 13 September 2023.
Markandya, A., & Wilkinson, P. (2007). Electricity generation and health. The Lancet, 370, 979–990.
Mokyr, J. (2012). Evolution and technological change: A new metaphor for economic history? Technological change (pp. 63–83). Routledge.
Nelson, R. R. (1985). An evolutionary theory of economic change. Harvard University Press.
Omohundro, S. (2008). The basic AI drives. Proceedings of the 2008 conference on Artificial General Intelligence 2008: Proceedings of the First AGI Conference, pp. 483–492. https://doi.org/10.5555/1566174
Park, P. S., Goldstein, S., O’Gara, A., Chen, M., & Hendrycks, D. (2023). AI deception: A survey of examples, risks, and potential solutions. arXiv preprint arXiv:2308.14752.
Rausand, M., Barros, A., & Høyland, A. (2020). System Reliability Theory: Models, Statistical Methods, and Applications (3rd ed.).
Richerson, P. J., & Boyd, R. (2006). Not by genes alone: How Culture Transformed Human evolution. University of Chicago Press.
Ruddiman, W. F. (2013). The anthropocene. Annual Review of Earth and Planetary Sciences, 41, 45–68.
Russell, S. (2019). Human compatible: Artificial intelligence and the problem of control. Penguin.
Schlaile, M. P., Mueller, M., Schramm, M., & Pyka, A. (2018). Evolutionary economics, responsible innovation and demand: Making a case for the role of consumers. Philosophy of Management, 17, 7–39.
Schlaile, M. P., Veit, W., & Boudry, M. (2023). Memes. In K. Dopfer, R. R. Nelson, J. Potts, & A. Pyka (Eds.), Routledge Handbook of Evolutionary Economics (pp. 235–248). Taylor & Francis.
Schumpeter, J., & Backhaus, U. (1934). The theory of economic development. Joseph Alois Schumpeter: Entrepreneurship, Style and Vision (pp. 61–116). Springer.
Stoop, J. (2017). How did aviation become so safe, and beyond? In: Proceedings of the 53rd ESReDA Seminar, 14–15 November 2017: European Commission Joint Research Centre, Ispra, Italy.
Suber, P. (2001). Saving Machines From Themselves: The Ethics of Deep Self-Modification. https://dash.harvard.edu/handle/1/32986888
Turner, A. (2021). A Meta-algorithm for the Collaborative Development of Artificial General Intelligence. https://bigmother.ai/resources/A_meta_algorithm_for_the_collaborative_development_of_Artificial_General_Intelligence-DRAFT-v02.pdf
World Nuclear Association (WNA) (2022). Safety of nuclear power reactors, World Nuclear Association, Safety of Nuclear Reactors - World Nuclear Association (world-nuclear.org), accessed 18 September 2023.
Zador, A., & LeCun, Y. (2019). Don’t fear the terminator. Scientific American. https://blogs.scientificamerican.com/observations/dont-fear-the-terminator/
Zou, A., Phan, L., Chen, S., Campbell, J., Guo, P., Ren, R., Pan, A., Yin, X., Mazeika, M., Dombrowski, A. K., Goel, S., Li, N., Byun, M. J., Wang, Z., Mallen, A., Basart, S., Koyejo, S., Song, D., Fredrikson, M., Kolter, J. Z., & Hendrycks, D. (2023). Representation engineering: A top-down approach to AI transparency. arXiv preprint arXiv:2310.01405.
Acknowledgements
We are grateful to Susan Blackmore, Andy Norman, Michael Schlaile, Steven Pinker, Cameron Domenico Kirk-Giannini, and an anonymous referee for discussions and helpful suggestions. We are especially grateful to Daniel Dennett (1942–2024), for being a tremendous source of inspiration and insight in evolutionary thinking, and for generously agreeing to discuss our paper in November 2023. We will miss him dearly.
Funding
The research of the corresponding author was partly funded by the Research Foundation - Flanders (FWO).
Author information
Authors and Affiliations
Contributions
Both authors contributed equally to the conception and writing process of this manuscript, and have read and approved the final version.
Corresponding author
Ethics declarations
Ethical approval and consent to participate
Not applicable.
Competing interests
The authors declare no conflicts of interest, financial or otherwise, that are directly or indirectly related to the work submitted for publication.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The original Online version of this article was revised: to update the position of an epigraph before introduction part.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Boudry, M., Friederich, S. The selfish machine? On the power and limitation of natural selection to understand the development of advanced AI. Philos Stud (2024). https://doi.org/10.1007/s11098-024-02226-3
Accepted:
Published:
DOI: https://doi.org/10.1007/s11098-024-02226-3