In recent years there has been a tremendous surge in the general capabilities of AI systems, mainly fuelled by training foundation models on internetscale data. Nevertheless, the creation of openended, ever self-improving AI remains elusive. In this position paper, we argue that the ingredients are now in place to achieve openendedness in AI systems with respect to a human observer. Furthermore, we claim that such open-endedness is an essential property of any artificial superhuman intelligence (ASI). We begin by providing a concrete formal definition of open-endedness through the lens of novelty and learnability. We then illustrate a path towards ASI via open-ended systems built on top of foundation models, capable of making novel, humanrelevant discoveries. We conclude by examining the safety implications of generally-capable openended AI. We expect that open-ended foundation models will prove to be an increasingly fertile and safety-critical area of research in the near future.
@article{arxiv.2406.04268,
title = {Open-Endedness is Essential for Artificial Superhuman Intelligence},
author = {Edward Hughes and Michael Dennis and Jack Parker-Holder and Feryal Behbahani and Aditi Mavalankar and Yuge Shi and Tom Schaul and Tim Rocktaschel},
journal= {arXiv preprint arXiv:2406.04268},
year = {2024}
}