what makes the historical record any less valid as a storage of brain structure? my impression is that any bit that depends on brain state is potentially a useful source of evidence, yeah? so eg getting a really low quality brain scan is potentially a useful sanity check for cryonic reconstruction, having lots of recordings of yourself muttering to yourself is a useful sanity check, and anything you ever did publicly would be compared to see if the life-trajectory implied by your reconstructed connectome seems to match correctly. perhaps you’d want to use those as a test set in cases where you have enough info from main reconstruction, but it’d still be quite useful, yeah?
My guess is this would be really underconstrained, maybe even if you had a ridiculous amount of data. The output you produce at some moment is a function of all past inputs. Brain state at that moment screens off past inputs, but then you’re trying to infer state at that moment with one piece of data, which is different from states at different moments.
Yeah, totally. I expect historical records and the memories of other people to be useful.
My point is that I don’t know an objective measure for whether the superintelligence rescued the existing person or built a new person, except via whether they match the other memories and records. If the superintelligence optimizes for the “rescued” person matching the memories of those who knew them, they will seem like they were revived successfully, but might not actually be very close to the real deal.
what makes the historical record any less valid as a storage of brain structure? my impression is that any bit that depends on brain state is potentially a useful source of evidence, yeah? so eg getting a really low quality brain scan is potentially a useful sanity check for cryonic reconstruction, having lots of recordings of yourself muttering to yourself is a useful sanity check, and anything you ever did publicly would be compared to see if the life-trajectory implied by your reconstructed connectome seems to match correctly. perhaps you’d want to use those as a test set in cases where you have enough info from main reconstruction, but it’d still be quite useful, yeah?
My guess is this would be really underconstrained, maybe even if you had a ridiculous amount of data. The output you produce at some moment is a function of all past inputs. Brain state at that moment screens off past inputs, but then you’re trying to infer state at that moment with one piece of data, which is different from states at different moments.
Yeah, totally. I expect historical records and the memories of other people to be useful.
My point is that I don’t know an objective measure for whether the superintelligence rescued the existing person or built a new person, except via whether they match the other memories and records. If the superintelligence optimizes for the “rescued” person matching the memories of those who knew them, they will seem like they were revived successfully, but might not actually be very close to the real deal.