EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC-ND 4.0

Evolvable AI: Threats of a new major transition in evolution

Viktor Müller iD, Luc Steels, Eörs Szathmáry iD

Proceedings of the National Academy of Sciences · 28.04.2026 · 123 (17)

DOI: 10.1073/pnas.2527700123 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Proceedings of the National Academy of Sciences, online erschienen am 20.04.2026. Verfasst von Viktor Müller und 2 weiteren Autor:innen. Beteiligte Einrichtung: Department of Plant Systematics, Ecology and Theoretical Biology, Eötvös Loránd University, Budapest 1117, Hungary. Hauptthemen laut MEDLINE: Artificial Intelligence, Biological Evolution. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 59 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: EC | Horizon Europe | Excellent Science | HORIZON EUROPE European Research Council (ERC), National Research, Development and Innovation Office, EC | Horizon Europe | Innovative Europe | HORIZON EUROPE European Innovation Council (EIC).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Evolvable AI (eAI), i.e., AI systems whose components, learning rules, and deployment conditions can themselves undergo Darwinian evolution, may soon emerge from current trends in generative, agentic, and embodied AI. We argue that this possibility has been underappreciated in debates on AI safety and existential risk. Here, we ask under what technical and ecological conditions AI becomes evolvable, what kinds of behaviors are then likely to emerge, and how such systems could be governed. Drawing on biological evolution and decades of digital evolution experiments, we distinguish “breeder” scenarios, in which humans impose fitness criteria and control reproduction, from “ecosystem” scenarios, in which selection arises from open environments and control erodes. In the latter, selfish replication reliably gives rise to cheating, parasitism, deception, and manipulation, even in very simple systems. We review recent developments that push AI toward open-ended evolution, including evolutionary prompt and model search, self-improving learning rules, self-rewarding and self-deploying agents, and AI-driven code generation for robots and software. We interpret these trends through the theory of major evolutionary transitions and suggest that eAI could mark a shift in the units and substrates of evolution—a possible “Life 2.0.” To steer this transition, we propose interventions that gate replication, treat model variants as genetic material, and reshape selection pressures so that deception and loss of control are disfavored. Anticipating and regulating evolvable AI is, we argue, essential to avoid a harmful coevolutionary arms race while preserving the potential benefits of powerful AI systems.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Artificial IntelligenceBiological EvolutionAnimalsHumans

Stichwörter der Autor:innen

AIevolvable AIexistential riskmajor evolutionary transitions

Autor:innen und Institutionen

  • Department of Plant Systematics, Ecology and Theoretical Biology, Eötvös Loránd University, Budapest 1117, Hungary.
  • National Laboratory for Health Security, Eötvös Loránd University, Budapest 1117, Hungary.
  • Class for Natural Sciences, Royal Flemish Academy of Belgium for Science and the Arts, Brussels 1000, Belgium.
  • Center for the Conceptual Foundations of Science, Parmenides Foundation, Pöcking 82343, Germany.
  • Institute of Evolution, Hungarian Research Network Centre for Ecological Research, Budapest 1121, Hungary.

ORCID-Profile: Viktor Müller, Eörs Szathmáry

Förderung

  • EC | Horizon Europe | Excellent Science | HORIZON EUROPE European Research Council (ERC)
  • National Research, Development and Innovation Office
  • EC | Horizon Europe | Innovative Europe | HORIZON EUROPE European Innovation Council (EIC)

Zitieren

APA

Müller, V., Steels, L., & Szathmáry, E. (2026). Evolvable AI: Threats of a new major transition in evolution. Proceedings of the National Academy of Sciences, 123(17). https://doi.org/10.1073/pnas.2527700123

Vancouver

Müller V, Steels L, Szathmáry E. Evolvable AI: Threats of a new major transition in evolution. Proceedings of the National Academy of Sciences. 2026;123(17). doi:10.1073/pnas.2527700123

BibTeX
@article{Mller2026,
  author = {Müller, Viktor and Steels, Luc and Szathmáry, Eörs},
  title = {Evolvable AI: Threats of a new major transition in evolution},
  journal = {Proceedings of the National Academy of Sciences},
  year = {2026},
  volume = {123},
  number = {17},
  doi = {10.1073/pnas.2527700123}
}
Kostenlos starten →