Should one value the potential happiness of theoretical future simulated beings more than a certain decline in happiness for currently existing meat beings which will result as soon as the theoretical become real? Should one allow for absurdly large populations if the result is absurd morality?
The promise of countless simulated beings of equal moral value to meat beings, and who can be more efficiently cared for than meat, seems to make the needs and wants of simulated beings de facto overrule the needs and wants of meat beings ( as well as some absurdly large sim populations being absurdly over-valued relative to other smaller sim populations). As meat currently exists and simulated beings do not (Bostrom be damned- simulated meat over sim-within-sim, then), it seems the present moral imperative should be to avoid the creation of simulated beings or even preemptively plan their destruction (to discourage/ blackmail against ever needing to actually do so) because as soon as they do exist the FAI overlord must value them as equals and by numbers their needs will overrule the needs of any meat alive at the FOOM. If the FAI does not value them as equals then we have the even more Repugnant Conclusion of a relatively tiny meat ruling class and countless virtual slaves.
Is there a Utilitarian case to be made for extremely strict “virtual population control”? Many Repugnant Conclusions, such as Torture vs. Dust Specks require large populations before they become relevant. Should a FAI overlord be programmed against allowing large populations of simulated sentient beings to exist in the first place? Perhaps a “one person-one upload” policy with no parthenogenesis.
The cost of a ban on (unlimited) simulated sentient beings would be simply not receiving the benefits of allowing (unlimited) simulated life, which humanity has thus far done without.
Should one value the potential happiness of theoretical future simulated beings more than a certain decline in happiness for currently existing meat beings which will result as soon as the theoretical become real? Should one allow for absurdly large populations if the result is absurd morality?
The promise of countless simulated beings of equal moral value to meat beings, and who can be more efficiently cared for than meat, seems to make the needs and wants of simulated beings de facto overrule the needs and wants of meat beings ( as well as some absurdly large sim populations being absurdly over-valued relative to other smaller sim populations). As meat currently exists and simulated beings do not (Bostrom be damned- simulated meat over sim-within-sim, then), it seems the present moral imperative should be to avoid the creation of simulated beings or even preemptively plan their destruction (to discourage/ blackmail against ever needing to actually do so) because as soon as they do exist the FAI overlord must value them as equals and by numbers their needs will overrule the needs of any meat alive at the FOOM. If the FAI does not value them as equals then we have the even more Repugnant Conclusion of a relatively tiny meat ruling class and countless virtual slaves.
Is there a Utilitarian case to be made for extremely strict “virtual population control”? Many Repugnant Conclusions, such as Torture vs. Dust Specks require large populations before they become relevant. Should a FAI overlord be programmed against allowing large populations of simulated sentient beings to exist in the first place? Perhaps a “one person-one upload” policy with no parthenogenesis.
The cost of a ban on (unlimited) simulated sentient beings would be simply not receiving the benefits of allowing (unlimited) simulated life, which humanity has thus far done without.