Are we both thinking of the book where vg gnxrf n qrhf rk znpuvan gb cerirag uhznavgl sebz qrfgeblvat vgfrys? Gur obbx va juvpu ng yrnfg bar punenpgre’f rkgencbyngrq ibyvgvba jbhyq cebonoyl qrfgebl uhznavgl, cnvashyyl?
When and which character? I’m not sure where you’re getting that.
Now the AI does seem absurd.
Well, it’s a pretty common error to think that with enough intelligence, an AI (or person) will be ethical and friendly. Eliezer himself made that mistake back in 2000 before he realized that intelligence is optimizing the world towards goals and those goals can be arbitrary. Spider was right that an AI would probably not have human emotions like greed or revenge, but he missed the idea that we’re made of atoms that the AI could use for something else.
Videotapes aren’t interactive, and it seems like the biggest benefit from the workshop would be actually engaging with the people and exercises involved.