Canaletto

Firstly, I'm assuming that high resolution human brain emulation that you can run on a computer is conscious in normal sense that we use in conversations. Like, it talks, has memories, makes new memories, have friends and hobbies and likes and dislikes and stuff. Just like a human that you could talk with only through videoconference type thing on a computer, but without actual meaty human on the other end. It would be VERY weird if this emulation exhibited all these human qualities for other reason than meaty humans exhibit them. Like, very extremely what the fuck surprising. Do you agree? So, we now have deterministic human file on our hands. Then, you can trivially make transformer like next token predictor out of human emulation. You just have emulation, then you feed it prompt (e.g. as appearing on piece of paper while they are sitting in a virtual room), then run it repeatedly for a constant time as it outputs each word adding that word to prompt (e.g. by recording all words that they say aloud). And artificially restrict it on 1000 words ingested. It would be a very deep model, but whatever. You can do many optimizations on this design, or special training/instructing to a human to behave more in line with the purpose of the setup Doesn't it suggest that this form factor of next token predictor isn't prohibitive for consciousness? Now, let's say there is human in time loop, that doesn't preserve their memories, completely resetting them every 30 minutes. No mater what they experience it doesn't persist. Is this human conscious? Well, yeah, duh. This human can talk, think, form memories on duration of these 30 minutes and have friends and likes and dislikes and stuff. But, would it be bad to hurt this human, hm? I think so, yeah. Most people probably agree. Now imagine you are going to be put in this time loop. You have 24 hours to coach yourself for it, and then the rest of the time you will be resetting to the state in the end of this preparatory tim