I gather that we’re talking about scientific information obtained through prior amoral acts, and that we’re not talking about the rather easier ethical question of a trade-off of current harm in exchange for good.
So again, if information obtained by prior amoral acts is the only basis for saving millions of innocent children, doubt that anyone would advocate that we should let them all die.
Then as Churchill said, we’ve established what you are and now we’re just haggling over the price.
Saving one innocent individual is surely also good, just less good than saving millions. So exactly how much current good is required before we can utilize scientific information that was obtained by means that caused prior harm?
Since we already have the data, presumably nobody is going to actually initiate an ethical research program to try to obtain the exact same data again. Should we really take some legalistic “fruit of the poisonous tree” stance here, should use of the data for good purposes depend on whether we can theoretically imagine whether we might in principle have obtained the same data through ethical research?