I’ve seen an argument that a nanotech organism with a reasonable level of error-correction could with high probability make error-free clones of itself until the heat death of the universe.
That seems plausible to me, but it’s still likely to be subject to selection because of competition for resources. Depending on its intelligence level and ethical structure, it might also be affected by arguments that it should limit its reproduction.
Offhand, I think they’d all include variation and selection.
I’ve seen an argument that a nanotech organism with a reasonable level of error-correction could with high probability make error-free clones of itself until the heat death of the universe.
That assumes the lack of black swans. Not a very good assumption when we extrapolate things until the heat death of the universe.
True, but it would have to be an exceedingly black swan to result in evolutionary-like mutations rather than simple annihilation.
First, annihilation is good enough—a destroyed nanobot fails at making “error-free clones of itself until the heat death of the universe”.
Second, all you need to do is to screw up the error-correction mechanism, the rest will take care of itself naturally.
That seems plausible to me, but it’s still likely to be subject to selection because of competition for resources. Depending on its intelligence level and ethical structure, it might also be affected by arguments that it should limit its reproduction.
The point is there’d be no variation for evolution to select on.
Why do you think there’d only be one sort of nanotech organism to select on and/or that perfect self-replication is the best or only strategy?
Well, the organism would need to be preprogrammed to survive in whatever environment it might find itself in until then.