Only if you submit an argument that involves a distinction. You keep making assumptions about what you have to submit that are not necessarily so. Your arguments do not have to be clever, or a stroke of genius/inspiration- or even be a total solution to the problem.
It just has to be the best submission you can make to the problem based on your philosophy. I just want people to be aware that trying to over-simplify the problem by pretending you both agree that X must die, so get drinking X will not get you anywhere. That, after all, is looking at the problem far too much as if there are clues left about how to distinguish AI from real, which is treating it like a logic puzzle again.
If you actually don't mind who dies, Azrael, then that works.
But... is that actually what you think? That is a 50/50 death chance, after all.
And yes, Castor's problem is not so much distingushing the AI, but thinking of a reason of how they both agree which one of them drinks- because right now either Castor A thinks he should drink, in which case Castor B thinks he (Castor B) should drink, or A thinks B should drink- in which case, B thinks A!
So every approach has its own problems.
it's based upon not being able to distinguish us. And me telling that to him. WE have to accept that we can't really decide who's the real Rade. So all we have to do is really both decide which one of us will drink. A or B. Both me and him agree on one...and then.....there's a 50% chance that our choice killed the real Rade....there's by bugger 😉