5 ms·
It would matter if true consciousness transfer turns out to actually be theoretically possible. First we have to figure out creating a near-perfect replica (giv
by meowface 7y ago
It would matter if true consciousness transfer turns out to actually be theoretically possible. First we have to figure out creating a near-perfect replica (give that a few centuries, at least, probably) and what substrates it could be implanted into, and then if it's possible to really retain subjective self-continuity. Non-transfered replicas will also each feel like they have continuity, so this is a difficult problem, but there's still a lot that needs to be understood before we can definitively rule the possibility in or out.
If the former is possible but not the latter, then I agree it isn't really very interesting. It's just a new way to spawn intelligent lifeforms, but we'll surely have many ways to do that by then. But I feel like if we can actually crack that, then we can eventually crack the second problem, too. The most popular hypothesis seems to be a Ship of Theseus-style gradual replacement / transition for some level of granularity (perhaps individual atoms, molecules, or neurons). We're fundamentally in the dark in so many ways at the moment that it seems unclear if these are plausible, obvious, or absurd.
Possible in the lifetimes of anyone reading this today? No (even if longevity breakthroughs accelerate like crazy). But if human or human-derived intelligence is still around in 1000, 100,000, 100,000,000 years from now, it's impossible to speculate what we may be capable of. If colonizing millions of galaxies becomes a boring walk in the park (via self-propagating autonomous constructs, for instance), then moving your consciousness to different tenancies while fully preserving your original self doesn't seem that insane. Maybe future "humans" will even skip the "middleman" (literally) and just be "born" directly into a non-biological housing from the start so that the transfer step won't be necessary, and maybe the "A" in AGI will become a non-sequitur.
- sooheon 7y agoKaibeezy linked to Greg Egan, who's one of my favorites in this area. To my mind the latter is logically impossible. Subjective self-continuity is not intrinsic to the experience of consciousness, it is a conclusion we draw upon introspection, based on circumstantial evidence. Subjectively, all we ever have is a sequence of fleeting moments. We only have the "word" of our memory to trust that the Now Me is a descendant of Past Me. As soon as perfect replication is possible, the replicant and the original have exactly equivalent claims to identity in every sense that matters, just one will have a better claim to physical continuity. But physical continuity is not the most important aspect of identity, it's just our intuitions favor it so heavily because it has historically perfectly coincided with conscious identity.
- blueadept111 7y agoRight, and we don't even have subjective continuity ourselves. Every night we fall asleep, and our consciousness dies. Every morning it is reborn. We don't worry about it.
- TomMarius 7y agoMaybe you are different than me, but I really don't feel as "reborn" when I wake up - I mean I immediately feel all the time that has passed (and how much, usually to 1 hour precision), and even during the night, when I'm sleeping, I'm still there, just in a different state and not making many memories, but I am definitely not gone.
- defertoreptar 7y agoA better example is going under anesthesia. I had a procedure done, and after I drifted to sleep, 30 minutes passed instantly and they were rolling me to the other room. The point is that we can lose our continuity of consciousness, yet we don't consider this "dying" in normal life. If we duplicate our exact physical brain and processes and immediately destroy the original while this happens, there is no loss of continuity, but this feels like dying still.
- TomMarius 7y agoLuckily I never had that done to me. To me, there is clear loss of continuity when you duplicate a brain. The process inside has been broken/does not continue. The structure is not the key, the key is the running state of the brain. Under anesthesia, the brain still functions.
- sooheon 7y agoI think you have to disambiguate two types of "breakage". One is suspending the thread, so to speak, and the other is functional destruction. I think perfectly non-destructive suspension of the process should not influence identity at all. Does an algorithm change its semantics if we pause the execution for a millisecond? This also assumes that duplication requires either type of breakage--for thought experiment purposes you may assume that cloning is possible with 0 delay and with 0 destructive change. If you care about which physical substrate the process is running on specifically, this is the bias towards physical continuity I noted.