Would it even be necessary for the EAI to threaten unbounded disutility? Given that U is unbounded in the positive direction as well, it seems like a simple threat by the EAI to cap U at some value would suffice. Depriving an agent of unbounded rewards could be as bad as threatening unbounded punishments. If the actions that the EAI wants the FAI to take do not themselves go against its utility function, then there is little reason for it not to comply, given the infinite rewards it can gain by going along. An infinite stick is persuasive, but so is an infinite carrot.
An upper bound to U′ (or U′′) could help to prevent threats of this sort.
In that case, one strategy the EAI might employ is to allow the FAI to increase its utility to an arbitrarily high level before threatening to take it away. In this way, it can simulate an arbitrarily large disutility even if the utility function is bounded below. Of course, a high utility might improve the FAI’s position to resist the EAI’s threats.
In this scenario, it is also possible that the FAI, anticipating the EAI’s future threat against it, might calculate its expected utility differently. For example, if it deduces that the EAI is waiting until some utility threshold b to make its threat, it might limit its own utility growth at some a<b if it found the threat credible to avoid triggering it.
This seems a lot like the human cognitive bias of loss aversion; I wonder if AGIs would (or should) suffer from something similar.
This doesn’t seem to me to work. It seems the FAI’s best reaction is simply to not grow above b until it’s sure the EAI can’t mess it up, rather than changing it’s own values to ensure the same result.
Would it even be necessary for the EAI to threaten unbounded disutility? Given that U is unbounded in the positive direction as well, it seems like a simple threat by the EAI to cap U at some value would suffice. Depriving an agent of unbounded rewards could be as bad as threatening unbounded punishments. If the actions that the EAI wants the FAI to take do not themselves go against its utility function, then there is little reason for it not to comply, given the infinite rewards it can gain by going along. An infinite stick is persuasive, but so is an infinite carrot.
An upper bound to U′ (or U′′) could help to prevent threats of this sort.
Ah, but we would like an infinite carrot :-) And cutting off U at the top would (almost) guarantee that we don’t get higher utility.
In that case, one strategy the EAI might employ is to allow the FAI to increase its utility to an arbitrarily high level before threatening to take it away. In this way, it can simulate an arbitrarily large disutility even if the utility function is bounded below. Of course, a high utility might improve the FAI’s position to resist the EAI’s threats.
In this scenario, it is also possible that the FAI, anticipating the EAI’s future threat against it, might calculate its expected utility differently. For example, if it deduces that the EAI is waiting until some utility threshold b to make its threat, it might limit its own utility growth at some a<b if it found the threat credible to avoid triggering it.
This seems a lot like the human cognitive bias of loss aversion; I wonder if AGIs would (or should) suffer from something similar.
This doesn’t seem to me to work. It seems the FAI’s best reaction is simply to not grow above b until it’s sure the EAI can’t mess it up, rather than changing it’s own values to ensure the same result.