The question whether reversible computation produces minds with relevant moral status is extremely important. Claude estimates me that it’d be a difference between having 1070 and 10124 mind-seconds instantiable in the reachable universe. (Because reversible minds could stretch our entropy budget long into the black hole era.)
Question is whether the reversing of the computation that makes up the mind and the lack of output (that’d imply bit-erasure) entail that the mind “didn’t really exist”.
There are four options here:
The mind didn’t exist because it’s just a computation+the computation being undone, no effect was made on the rest of the universe.
The mind existed once, during the “forward pass”.
The mind existed twice, during the “forward pass”, and during the reversion, since they’re isomorphic very similar computations.
There were two different, very similar minds.
Maybe I’ll take a longer stab at this at some point, but looks like nobody has thought about this except Tomasik 2015 mentioning it in passing.
3. The mind existed twice, during the “forward pass”, and during the reversion, since they’re isomorphic computations.
They’re not isomorphic. “Dual” would be a better word, or you could just say that one is a mirror image/inversion of the other.
In which case, the question is: Does a sequence of computations that compose up to a morally-relevant-mind-moment also compose up to a morally-relevant-mind-moment if you compose their inverses?
MR(◯ni=1fi)=⊤?⟹MR(◯1i=nf−1i)=⊤
I don’t have an informed view here, but it is the case that reality doesn’t always treat dual pairs equally. E.g., I recall davidad saying somewhere that he preferred using colimits instead of limits for modeling things (even though limits would also work) because they’re computationally more tractable[1] or something (in the context of world modelling for the Safeguarded AI program).
Which is to say, it’s not immediately obvious that a chain of reversible operations that composes up to a mind-process will also compose up to a mind-process if you invert it.
It would actually be a kind of fun situation if computation that is the inversion of a reversible computation that would normally instantiate a stream of experience, would “undo” those experiences (in the morally relevant sense or whatever other relevant sense).
Which makes sense on first glance because colimits generalize disjoint sums (which add cardinalities: |A⊔B|=|A|+|B|), whereas limits generalize cartesian products (which multiply cardinalities: |A×B|=|A|⋅|B|).
if it’s just about whether there was no effect, you could totally reversibly compute 10124−70 minds for one second, summarize their outputs into one bit (e.g. whether most of them voted in favor of a proposal), observe that bit, and then reverse the mind-computation to the initial state.
Yeah… seems right. I could cop out with “ah, the number of bit erasures by a program matters, not just whether it did”, but I don’t have any good reason for believing this.
Off the top of my head, it feels like this might be somehow connected to the idea of logical zombies (I don’t have a clear sense of what the connection is exactly; will think more about it later).
You also made me wonder what a shortlist of crucial considerations for digital minds ranked by something like instantiable mind-seconds “swing factor” (or whatever the more sophisticated version should be) would look like, where reversible computation’s swing factor is 10^54 by Claude’s lights.
How to aggregate minds being computed in quantum superposition—do our utilities need to have amplitudes too? (Claude tells me that philosophers haven’t thought about this which seems absurd to me)
People have thought about it but I also don’t know of anything extensive being written. Counting discretely if a consciousnesses “exists once” or “exists twice” don’t seem like the right take to me. What happens if you cut the circuits in your CPU into 10 pieces that each do the same computation independently—do you get 10 times as many conscious beings? (A possible solution to this puzzle is that the “anthropic measure” of a computation has a multiplicative factor by the number of bits the computation erases, which would mean that reversible computers aren’t conscious)
[EDIT] I’ll mention that reversible computations, even if not part of a quantum computer, are quantumly weird. We’re not sure about much about anthropics, but one thing we are confident in is the Born rule. Whatever your interpretation of quantum mechanics is, it probably involves decoherence. And reversible computations don’t decohere.
Comment from 12 years ago, −3 karma… that’s a deep cut. It is related to the thick-wire problem, yeah, good point.
I’m not super convinced (but intrigued) by your proposal that a computation is not conscious unless it erases bits.
Additional thought: If we accept UDASSA or some other computational view of anthropics+cosmology then we are (mostly, in terms of measure) computations embedded in a bigger, reversible, computation. Maybe it’s about how many output bits of the universe-computation I affect, and if I’m reversible I don’t affect any?
The quantum connection feels weird, and I don’t feel like I understand this stuff well enough to comment, but at least Claude Sonnet 4.5 tells me philosophers haven’t yet argued about the moral weight of minds in superposition, which is really surprising to me.
In case I don’t write a full post about this:
The question whether reversible computation produces minds with relevant moral status is extremely important. Claude estimates me that it’d be a difference between having 1070 and 10124 mind-seconds instantiable in the reachable universe. (Because reversible minds could stretch our entropy budget long into the black hole era.)
Question is whether the reversing of the computation that makes up the mind and the lack of output (that’d imply bit-erasure) entail that the mind “didn’t really exist”.
There are four options here:
The mind didn’t exist because it’s just a computation+the computation being undone, no effect was made on the rest of the universe.
The mind existed once, during the “forward pass”.
The mind existed twice, during the “forward pass”, and during the reversion, since they’re
isomorphicvery similar computations.There were two different, very similar minds.
Maybe I’ll take a longer stab at this at some point, but looks like nobody has thought about this except Tomasik 2015 mentioning it in passing.
They’re not isomorphic. “Dual” would be a better word, or you could just say that one is a mirror image/inversion of the other.
In which case, the question is: Does a sequence of computations that compose up to a morally-relevant-mind-moment also compose up to a morally-relevant-mind-moment if you compose their inverses?
MR(◯ni=1fi)=⊤?⟹MR(◯1i=nf−1i)=⊤
I don’t have an informed view here, but it is the case that reality doesn’t always treat dual pairs equally. E.g., I recall davidad saying somewhere that he preferred using colimits instead of limits for modeling things (even though limits would also work) because they’re computationally more tractable[1] or something (in the context of world modelling for the Safeguarded AI program).
Which is to say, it’s not immediately obvious that a chain of reversible operations that composes up to a mind-process will also compose up to a mind-process if you invert it.
It would actually be a kind of fun situation if computation that is the inversion of a reversible computation that would normally instantiate a stream of experience, would “undo” those experiences (in the morally relevant sense or whatever other relevant sense).
Which makes sense on first glance because colimits generalize disjoint sums (which add cardinalities: |A⊔B|=|A|+|B|), whereas limits generalize cartesian products (which multiply cardinalities: |A×B|=|A|⋅|B|).
if it’s just about whether there was no effect, you could totally reversibly compute 10124−70 minds for one second, summarize their outputs into one bit (e.g. whether most of them voted in favor of a proposal), observe that bit, and then reverse the mind-computation to the initial state.
Yeah… seems right. I could cop out with “ah, the number of bit erasures by a program matters, not just whether it did”, but I don’t have any good reason for believing this.
Off the top of my head, it feels like this might be somehow connected to the idea of logical zombies (I don’t have a clear sense of what the connection is exactly; will think more about it later).
Yes please to the longer stab.
You also made me wonder what a shortlist of crucial considerations for digital minds ranked by something like instantiable mind-seconds “swing factor” (or whatever the more sophisticated version should be) would look like, where reversible computation’s swing factor is 10^54 by Claude’s lights.
Argh too many possible projects :-D
Other possible crucial considerations:
How to aggregate minds being computed in quantum superposition—do our utilities need to have amplitudes too? (Claude tells me that philosophers haven’t thought about this which seems absurd to me)
How to aggregate minds being computed in neural network superposition? This one may be simpler.
Malament-Hogarth spacetimes? If feasible then WTF?
I think you’re looking for the concept of sensitivity analysis for a credence.
People have thought about it but I also don’t know of anything extensive being written. Counting discretely if a consciousnesses “exists once” or “exists twice” don’t seem like the right take to me. What happens if you cut the circuits in your CPU into 10 pieces that each do the same computation independently—do you get 10 times as many conscious beings? (A possible solution to this puzzle is that the “anthropic measure” of a computation has a multiplicative factor by the number of bits the computation erases, which would mean that reversible computers aren’t conscious)
[EDIT] I’ll mention that reversible computations, even if not part of a quantum computer, are quantumly weird. We’re not sure about much about anthropics, but one thing we are confident in is the Born rule. Whatever your interpretation of quantum mechanics is, it probably involves decoherence. And reversible computations don’t decohere.
Comment from 12 years ago, −3 karma… that’s a deep cut. It is related to the thick-wire problem, yeah, good point.
I’m not super convinced (but intrigued) by your proposal that a computation is not conscious unless it erases bits.
Additional thought: If we accept UDASSA or some other computational view of anthropics+cosmology then we are (mostly, in terms of measure) computations embedded in a bigger, reversible, computation. Maybe it’s about how many output bits of the universe-computation I affect, and if I’m reversible I don’t affect any?
The quantum connection feels weird, and I don’t feel like I understand this stuff well enough to comment, but at least Claude Sonnet 4.5 tells me philosophers haven’t yet argued about the moral weight of minds in superposition, which is really surprising to me.
I think you meant to say “is not conscious”.