All three of these are hard, and all three fail catastrophically.
I would be very surprised if all three of these are equally hard, and I suspect that (1) is the easiest and by a long shot.
Making a human imitator AI, once you already have weakly superhuman AI is a matter of cutting down capabilities and I suspect that it can be achieved by distillation, i.e. using the weakly superhuman AI that we will soon have to make a controlled synthetic dataset for pretraining and finetuning and then a much larger and more thorough RLHF dataset.
Finally you’d need to make sure the model didn’t have too many parameters.
I would be very surprised if all three of these are equally hard, and I suspect that (1) is the easiest and by a long shot.
Making a human imitator AI, once you already have weakly superhuman AI is a matter of cutting down capabilities and I suspect that it can be achieved by distillation, i.e. using the weakly superhuman AI that we will soon have to make a controlled synthetic dataset for pretraining and finetuning and then a much larger and more thorough RLHF dataset.
Finally you’d need to make sure the model didn’t have too many parameters.