A couple ways to implement a hybrid approach with existing AI safety tools:
Logical Induction: Specify some computationally expensive simulation of idealized humans. Run a logical inductor with the deductive process running the simulation and outputting what the humans say after time x in simulation, as well as statements about what non-idealized humans are saying in the real world. The inductor should be able to provide beliefs about what the idealized humans will say in the future informed by information from the non-idealized humans.
HCH/IDA: The HCH-humans demonstrate a reasoning process which aims to predict the output of a set of idealized humans using all available information (which can include running simulations of idealized humans or information from real humans). The way that the HCH tree using information about real humans involves looking carefully at their circumstances and asking things like “how do the real human’s circumstances differ from the idealized human” and “is the information from the real human compromised in some way?”
A couple ways to implement a hybrid approach with existing AI safety tools:
Logical Induction: Specify some computationally expensive simulation of idealized humans. Run a logical inductor with the deductive process running the simulation and outputting what the humans say after time x in simulation, as well as statements about what non-idealized humans are saying in the real world. The inductor should be able to provide beliefs about what the idealized humans will say in the future informed by information from the non-idealized humans.
HCH/IDA: The HCH-humans demonstrate a reasoning process which aims to predict the output of a set of idealized humans using all available information (which can include running simulations of idealized humans or information from real humans). The way that the HCH tree using information about real humans involves looking carefully at their circumstances and asking things like “how do the real human’s circumstances differ from the idealized human” and “is the information from the real human compromised in some way?”