Skip to main content

A Computability Argument Against Superintelligence

“It turns out that, yes, there are limits to computations based on the laws of physics. But these still allow for a continuation of exponential growth until non-biological intelligence is trillion of trillions times more powerful than all of human civilization today, contemporary computers included.”

Ray Kurzweil, in “The Singularity is Near: When Humans Transcend Biology”

Abstract

Using the contemporary view of computing exemplified by recent models and results from non-uniform complexity theory, we investigate the computational power of cognitive systems. We show that in accordance with the so-called extended Turing machine paradigm such systems can be modelled as non-uniform evolving interactive systems whose computational power surpasses that of the classical Turing machines. Our results show that there is an infinite hierarchy of cognitive systems. Within this hierarchy, there are systems achieving and surpassing the human intelligence level. Any intelligence level surpassing the human intelligence is called the superintelligence level. We will argue that, formally, from a computation viewpoint the human-level intelligence is upper-bounded by the \(\Upsigma_2\) class of the Arithmetical Hierarchy. In this class, there are problems whose complexity grows faster than any computable function and, therefore, not even exponential growth of computational power can help in solving such problems, or reach the level of superintelligence.

This is a preview of subscription content, access via your institution.

Notes

  1. 1.

    Arithmetical Hierarchy is the hierarchy of unsolvable problems of increasing computational difficulty. The respective problems are defined with the help of certain sets based on the complexity of quantified logic formulas that define them (cf. [6]).

References

  1. 1.

    Balcázar JL, Diáz J, Gábarró J. Structural complexity, Vol. I, 2nd ed. Berlin: Springer; 1995.

    Book  Google Scholar 

  2. 2.

    Bernstain E, Vazirani U. Quantum complexity theory. In: Proceeding of the 25th annual symposium on the theory of computing. New York: ACM; 1993. p. 11–20.

  3. 3.

    Blum M. Can (theoretical computer) science come to grips with consciousness? Invited talk at FOCS 2009, 2009.

  4. 4.

    Chalmers DJ. A computational foundation for the study of cognition. Minds Mach. 1994;4(4).

  5. 5.

    Chalmers DJ. The singularity: a philosophical analysis, 2009, cf. http://consc.net/papers/singularity.pdf.

  6. 6.

    Cooper SB. Computability theory. London: Chapman & Hall/CRC; 2004.

    Google Scholar 

  7. 7.

    Cooper SB. From descartes to Turing: the computational content of supervenience. In: Burgin M, Dodig-Crnkovic G, editors. Information and computation. Singapore: World Scientific Publishing Co; 2011. p. 107–48.

    Google Scholar 

  8. 8.

    Deutsch D. Quantum theory, the Church-Turing principle and the universal quantum computer. Proc R Soc Lond A 1985;400:97–117.

    Article  Google Scholar 

  9. 9.

    Étesi G, Nésmeti I. Non-Turing computations via Malament–Hogarth space-times. Int. J. Theor. Phys. 2002;41:341–70.

    Article  Google Scholar 

  10. 10.

    Goertzel B. Artificial general intelligence: now is the time. Published on KurzweilAI.net, April 9, 2007.

  11. 11.

    Goldin DQ, Smolka SA, Attie PC, Sonderegger E. Turing machines, transition systems, and interaction. Inf Comput 2004;194(2):101–28.

    Article  Google Scholar 

  12. 12.

    Green MW. A lower bound on Rado’s sigma function for binary Turing machines. In: Proceedings of the IEEE 5th annual symposium on switching circuits theory and logical design; 1964. p. 91–4

  13. 13.

    Hopcroft J, Motwani R, Ullman JD. Introduction to automata theory, languages and computation, 2nd ed. Reading: Addison-Wesley; 2000.

    Google Scholar 

  14. 14.

    Karp RM, Lipton RJ. Some connections between non-uniform and uniform complexity classes. In: Proceeding 12th annual ACM symposium on the theory of computing (STOC’80); 1980. p. 302–9.

  15. 15.

    Karp RM, Lipton R. Turing machines that take advice, L’Enseignement Mathématique, IIe Série, Tome XXVIII; 1982. p. 191–209.

  16. 16.

    Knuth’s up-arrow notation at http://en.wikipedia.org/wiki/Knuth_up-arrow_notation.

  17. 17.

    Kurzweil, R.: The singularity is near. New York: Viking Books; 2005. p. 652.

    Google Scholar 

  18. 18.

    Levin LA. Universal sequential search problems. Probl Inf Transm 1973;9(3):265–66.

    Google Scholar 

  19. 19.

    Lloyd S. Ultimate physical limits to computation. Nature 2000;406(6799):1047–54.

    PubMed  Article  CAS  Google Scholar 

  20. 20.

    Lloyd S. How fast, how small, and how powerful: Moor’s law and the ultimate laptop. In: Rebooting civilization, http://www.edge.org/3rd_culture/rebooting/rebooting.html, 2001.

  21. 21.

    Michel P. Historical survey of busy beavers, http://www.logique.jussieu.fr/michel/ha.html#tm62.

  22. 22.

    Penrose R. Shadows of the mind (a search for the missing science of consciousness). Oxford: Oxford University Press; 1994. p. 457

    Google Scholar 

  23. 23.

    Shettleworth S. Cognition, evolution, and behavior. Oxford: Oxford University Press; 1998.

    Google Scholar 

  24. 24.

    Turing AM. On computable numbers, with an application to the Entscheidungsproblem. Proc London Math Soc. 1936;42-2:230–65. A correction, ibid. 1937;43:544–46.

  25. 25.

    Turing AM. Systems of logic based on ordinals. Proc Lond Math Soc Ser 2, 1939;45:161–228.

    Article  Google Scholar 

  26. 26.

    van Leeuwen J, Wiedermann J. The Turing machine paradigm in contemporary computing. In: Enquist B, Schmidt W, editors, Mathematics unlimited—2001 and beyond. Berlin: Springer; 2001. p. 1139–55.

    Chapter  Google Scholar 

  27. 27.

    van Leeuwen J, Wiedermann J. Beyond the Turing limit: evolving interactive systems. In: Proceeding SOFSEM’01, LNCS Vol. 2234. Berlin: Springer; 2001. p. 90–109.

  28. 28.

    van Leeuwen J, Wiedermann J. A theory of interactive computation. In: Goldin D, Smolka SA, Wegner P, editors. Interactive computation: the new paradigm. Berlin: Springer; 2006. p. 119–42.

    Chapter  Google Scholar 

  29. 29.

    van Leeuwen J, Wiedermann J. Computation as unbounded process. Theoretical computer science; 2012. doi:10.1016/j.tcs2011.12.049.

  30. 30.

    Verbaan PRA, van Leeuwen J, Wiedermann J. Complexity of evolving interactive systems. In: Karhumäki J et al editors, Theory is forever, essays dedicated to arto Salomaa on the occasion of his 70th birthday. LNCS Vol. 3113. Berlin: Springer; 2004, p. 268–81.

  31. 31.

    Verbaan PRA. The computational complexity of evolving systems, Ph.D. Thesis, Dept. of Information and Computing Sciences, Utrecht University; 2006.

  32. 32.

    Wiedermann J, van Leeuwen J. The emergent computational potential of evolving artificial living systems. Ai Commun 2002;15(4):205–16.

    Google Scholar 

  33. 33.

    Wiedermann J, van Leeuwen J. Relativistic computers and non-uniform complexity theory. In: Unconvential models of computation (UMC’2002), LNCS Vol. 2509. Berlin: Springer; 2002, p. 287–99.

  34. 34.

    Wiedermann J, van Leeuwen J. How we think of computing today. (invited talk). In: Proceeding CiE 2008, LNCS 5028. Berlin: Springer; 2008. p. 579–93.

Download references

Acknowledgments

This research was carried out within the institutional research plan AV0Z10300504 and partially supported by a GA ČR grant No. P202/10/1333

Author information

Affiliations

Authors

Corresponding author

Correspondence to Jiří Wiedermann.

Additional information

A preliminary version of this paper appeared in Proceedings of the AISB’11—Computing and Philosophy, The University of York, York, UK, April 2011, pages 73–79.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Wiedermann, J. A Computability Argument Against Superintelligence. Cogn Comput 4, 236–245 (2012). https://doi.org/10.1007/s12559-012-9124-9

Download citation

Keywords

  • Cognitive systems
  • Intelligence
  • Extended Turing machine thesis
  • Singularity
  • Superintelligence