Towards shutdownable agents via stochastic choice

Elliott Thornley (Global Priorities Institute, University of Oxford), Alexander Roman (New College of Florida), Christos Ziakas (Imperial College, London), Leyton Ho (Brown University) and Louis Thomson (University of Oxford)

GPI Working Paper No. 16-2024

The Incomplete Preferences Proposal (IPP) is an idea for ensuring that advanced artificial agents never resist shutdown. 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 provide some initial evidence that DReST reward functions could train advanced agents to be USEFUL and NEUTRAL. Our theoretical work suggests that these agents would be useful and shutdownable.

Other working papers

The Shutdown Problem: An AI Engineering Puzzle for Decision Theorists – Elliott Thornley (Global Priorities Institute, University of Oxford)

I explain and motivate the shutdown problem: the problem of designing artificial agents that (1) shut down when a shutdown button is pressed, (2) don’t try to prevent or cause the pressing of the shutdown button, and (3) otherwise pursue goals competently. I prove three theorems that make the difficulty precise. These theorems suggest that agents satisfying some innocuous-seeming conditions will often try to prevent or cause the pressing of the shutdown button, even in cases where it’s costly to do so. I end by noting that…

The evidentialist’s wager – William MacAskill, Aron Vallinder (Global Priorities Institute, Oxford University) Caspar Österheld (Duke University), Carl Shulman (Future of Humanity Institute, Oxford University), Johannes Treutlein (TU Berlin)

Suppose that an altruistic and morally motivated agent who is uncertain between evidential decision theory (EDT) and causal decision theory (CDT) finds herself in a situation in which the two theories give conflicting verdicts. We argue that even if she has significantly higher credence in CDT, she should nevertheless act …

High risk, low reward: A challenge to the astronomical value of existential risk mitigation – David Thorstad (Global Priorities Institute, University of Oxford)

Many philosophers defend two claims: the astronomical value thesis that it is astronomically important to mitigate existential risks to humanity, and existential risk pessimism, the claim that humanity faces high levels of existential risk. It is natural to think that existential risk pessimism supports the astronomical value thesis. In this paper, I argue that precisely the opposite is true. Across a range of assumptions, existential risk pessimism significantly reduces the value of existential risk mitigation…