Imperfect Recall and AI Delegation
Eric Olav Chen (Global Priorities Institute, University of Oxford), Alexis Ghersengorin (Global Priorities Institute, University of Oxford) and Sami Petersen (Department of Economics, University of Oxford)
GPI Working Paper No. 30-2024
A principal wants to deploy an artificial intelligence (AI) system to perform some task. But the AI may be misaligned and pursue a conflicting objective. The principal cannot restrict its options or deliver punishments. Instead, the principal can (i) simulate the task in a testing environment and (ii) impose imperfect recall on the AI, obscuring whether the task being performed is real or part of a test. By committing to a testing mechanism, the principal can screen the misaligned AI during testing and discipline its behaviour in deployment. Increasing the number of tests allows the principal to screen or discipline arbitrarily well. The screening effect is preserved even if the principal cannot commit or if the agent observes information partially revealing the nature of the task. Without commitment, imperfect recall is necessary for testing to be helpful.
Other working papers
A non-identity dilemma for person-affecting views – Elliott Thornley (Global Priorities Institute, University of Oxford)
Person-affecting views in population ethics state that (in cases where all else is equal) we’re permitted but not required to create people who would enjoy good lives. In this paper, I present an argument against every possible variety of person- affecting view. The argument takes the form of a dilemma. Narrow person-affecting views must embrace at least one of three implausible verdicts in a case that I call ‘Expanded Non- Identity.’ Wide person-affecting views run into trouble in a case that I call ‘Two-Shot Non-Identity.’ …
It Only Takes One: The Psychology of Unilateral Decisions – Joshua Lewis (New York University) et al.
Sometimes, one decision can guarantee that a risky event will happen. For instance, it only took one team of researchers to synthesize and publish the horsepox genome, thus imposing its publication even though other researchers might have refrained for biosecurity reasons. We examine cases where everybody who can impose a given event has the same goal but different information about whether the event furthers that goal. …
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)
Some worry that advanced artificial agents may resist being shut down. The Incomplete Preferences Proposal (IPP) is an idea for ensuring that does not 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…