Another type might be capability elicitation, e.g. try to train a model organism that has a dangerous capability to see if it’s even possible / how hard it is to train a current model to have that trait.
Yes, good point. Something like gain of function research. Could be useful for predicting / forecasting future model capabilities.
Another type might be capability elicitation, e.g. try to train a model organism that has a dangerous capability to see if it’s even possible / how hard it is to train a current model to have that trait.
Yes, good point. Something like gain of function research. Could be useful for predicting / forecasting future model capabilities.