Towards shutdownable agents via stochastic choice
Elliott Thornley (Global Priorities Institute, University of Oxford), Alexander Roman (New College of Florida), Christos Ziakas (Independent), Leyton Ho (Brown University) and Louis Thomson (University of Oxford)
GPI Working Paper No. 16-2024
Some worry that advanced artificial agents may resist being shut down. The Incomplete Preferences Proposal (IPP) is an idea for ensuring that doesn’t happen. A key part of the IPP is using a novel ‘Discounted REward for Same-Length Trajectories (DREST)’ reward function to train agents to (1) pursue goals effectively conditional on each trajectory-length (be ‘USEFUL’), and (2) choose stochastically between different trajectory-lengths (be ‘NEUTRAL’ about trajectory-lengths). In this paper, we propose evaluation metrics for USEFULNESS and NEUTRALITY. We use a DREST reward function to train simple agents to navigate gridworlds, and we find that these agents learn to be USEFUL and NEUTRAL. Our results thus suggest that DREST reward functions could also train advanced agents to be USEFUL and NEUTRAL, and thereby make these advanced agents useful and shutdownable.
Other working papers
Desire-Fulfilment and Consciousness – Andreas Mogensen (Global Priorities Institute, University of Oxford)
I show that there are good reasons to think that some individuals without any capacity for consciousness should be counted as welfare subjects, assuming that desire-fulfilment is a welfare good and that any individuals who can accrue welfare goods are welfare subjects. While other philosophers have argued for similar conclusions, I show that they have done so by relying on a simplistic understanding of the desire-fulfilment theory. My argument is intended to be sensitive to the complexities and nuances of contemporary…
AI takeover and human disempowerment – Adam Bales (Global Priorities Institute, University of Oxford)
Some take seriously the possibility of AI takeover, where AI systems seize power in a way that leads to human disempowerment. Assessing the likelihood of takeover requires answering empirical questions about the future of AI technologies and the context in which AI will operate. In many cases, philosophers are poorly placed to answer these questions. However, some prior questions are more amenable to philosophical techniques. What does it mean to speak of AI empowerment and human disempowerment? …
Is Existential Risk Mitigation Uniquely Cost-Effective? Not in Standard Population Models – Gustav Alexandrie (Global Priorities Institute, University of Oxford) and Maya Eden (Brandeis University)
What socially beneficial causes should philanthropists prioritize if they give equal ethical weight to the welfare of current and future generations? Many have argued that, because human extinction would result in a permanent loss of all future generations, extinction risk mitigation should be the top priority given this impartial stance. Using standard models of population dynamics, we challenge this conclusion. We first introduce a theoretical framework for quantifying undiscounted cost-effectiveness over…