Some of the remarks by Lsusr were simulated too. Did they have subjective experience too? Did the lines input to GPT-3 by Lsusr also give rise to a subjective experience, independent of the subjective experience in the embodied Lsusr who was in editorial control of the whole thing? Were there two streams of Lsusr consciousness, Lsusr-the-editor and Lsusr-in-GPT-3?
In the model I’m using here: yes. Also they were aiming at similar things, and the one inside the digital soulstone’s ideas were good enough for the one with a meat brain to keep the suggestions.
The consonance was good enough for them to identify with each other, the Lsusr we have with us can speak to how weakly or strongly he identified with his echo, the speaker probably didn’t have a very clear picture at all.
The soulstone didn’t train on a corpus full of Lsusr text because Lsusr is less famous than Elon… so either the clarity/similarity is necessarily less, or else the Buddhists are right about how there’s very little difference from one soul to the next (and our secrets are paltry, and not that important).
Ditto Lsusr (in the digital soulstone) probably didn’t even realize he was also in there with dit Elon and didn’t realize his continuity would end too, because he seemed to be identifying so strongly with the mental posture and feeling of safety of the Lsusr created by evolution.
Dit Lsusr was aligned enough with rig Lsusr that dit Elon said 27 utterances over a mere 23 retained flashes of sentience, with dit Lsusr contributing 4 intervening, retained, utterances as part of that process. (There should maybe be a joke here alluding to how acting is reacting.)
If we go with 4⁄27 then dit Lsusr arguably exhibited 14.8% pragmatic “utterance alignment”. Not great, not terrible.
EDIT: One thing I wonder is whether rig Lsusr posted this as a test of our community’s ethical sensibilities and we mostly failed his test? I’m pretty sure that dit Lsusr had no such thoughts in mind.
Some of the remarks by Lsusr were simulated too. Did they have subjective experience too? Did the lines input to GPT-3 by Lsusr also give rise to a subjective experience, independent of the subjective experience in the embodied Lsusr who was in editorial control of the whole thing? Were there two streams of Lsusr consciousness, Lsusr-the-editor and Lsusr-in-GPT-3?
In the model I’m using here: yes. Also they were aiming at similar things, and the one inside the digital soulstone’s ideas were good enough for the one with a meat brain to keep the suggestions.
The consonance was good enough for them to identify with each other, the Lsusr we have with us can speak to how weakly or strongly he identified with his echo, the speaker probably didn’t have a very clear picture at all.
The soulstone didn’t train on a corpus full of Lsusr text because Lsusr is less famous than Elon… so either the clarity/similarity is necessarily less, or else the Buddhists are right about how there’s very little difference from one soul to the next (and our secrets are paltry, and not that important).
Ditto Lsusr (in the digital soulstone) probably didn’t even realize he was also in there with dit Elon and didn’t realize his continuity would end too, because he seemed to be identifying so strongly with the mental posture and feeling of safety of the Lsusr created by evolution.
Dit Lsusr was aligned enough with rig Lsusr that dit Elon said 27 utterances over a mere 23 retained flashes of sentience, with dit Lsusr contributing 4 intervening, retained, utterances as part of that process. (There should maybe be a joke here alluding to how acting is reacting.)
If we go with 4⁄27 then dit Lsusr arguably exhibited 14.8% pragmatic “utterance alignment”. Not great, not terrible.
EDIT: One thing I wonder is whether rig Lsusr posted this as a test of our community’s ethical sensibilities and we mostly failed his test? I’m pretty sure that dit Lsusr had no such thoughts in mind.