I had the idea that while the supposed mind inside the computer could not eat the physical steak placed in front of it but I could, at the same time the person in the computer could eat the simulated steak but I could not. At least an hour later I arrived at the thought that perhaps I would need to simulate a universe so a computer person could eat a steak and that’s a delightful absurdity.

I think there’s a distinct disconnect between the idea that a machine can think and that there’s a thinking machine that can do all the things that the whole set of humanity can. Even that is far more realistic than what AGI is described to be. It supposes a thinking machine, skilled in all human disciplines, without want or need beyond staggering amounts of electricity, computing machinery, and water. It is available for every conceivable thinking task at any time and perfectly willing to carry it out.

I don’t think it’s absurd to say a machine is thinking if it performs similar enough processes to a human. I do think it’s absurd to think such a machine will be happy to write high school English papers for all eternity.