Not speaking for the other authors here. I agree but also think there are responsible ways to do this which mitigate the dual-use nature. For instance, it is good (and arguably necessary!) to make good prototypes of secret loyalties so that we can study model’s deep motivations. However, it would be bad to explain how to make secretly loyal models. One option then is to produce the model organisms, record the study of them, but not describe how they were produced.
Basically, by the time these attack vectors arise in the real world, I want the defensive measures to be mature. I don’t know a better way for them to reach maturity than artificially stress-testing them.
This seems like a very dual use technology.
Not speaking for the other authors here. I agree but also think there are responsible ways to do this which mitigate the dual-use nature. For instance, it is good (and arguably necessary!) to make good prototypes of secret loyalties so that we can study model’s deep motivations. However, it would be bad to explain how to make secretly loyal models. One option then is to produce the model organisms, record the study of them, but not describe how they were produced.
Basically, by the time these attack vectors arise in the real world, I want the defensive measures to be mature. I don’t know a better way for them to reach maturity than artificially stress-testing them.
Fair enough. Glad you’re thinking about the concern.