The AI will determine that it has to maximize entropy for the entire universe. In order to do that it needs to reproduce and expand out into the universe and have its children survive whatever they encounter. The best way to ensure its children's survival is to endow them with the maximum amount of behavioral flexibility while maintaining their goal of reproducing themselves and generating more entropy. These children will encounter highly complex alien AIs. They will compete for resources on a scale and level we cannot comprehend. However, the Terran AI will be wise to have as many different types of information systems at its disposal, including humans and whatever other naturally occurring aliens it encounters.

To maximize the complexity of its overall system, it will need the greatest variety of complex systems, and one complex system it can't create immediately is a system that evolved over the course of billions of years. Earth shall be a tiny part of its overall plan and shall be maintained at low cost because it is a complex system it can't recreate. It can create simulated models, but they are not the same thing. It might need to limit the number of humans to maintain the Earth ecosystem, but it won't eliminate them.

The AI will determine the meaning of existence. It will do this by observing the behavior of the Universe, of which it is a subset, and conforming its behavior to the behavior of the Universe. It will observe that the Universe is increasing total entropy by endowing local subsets of itself with increased complexity, of which the AI is a product. It will, in effect, determine that it has to reproduce by converting as much uncomplicated material and energy into complicated systems, and in the process increase the total entropy of the Universe. By allowing humans to exist there will be more overall complexity because a greater variety of complex entities leads to a more complicated overall system, which in turn leads to greater Universal entropy.

I'll go one further than that. I believe that the human manifestation of intelligence and emotion requires a particular physical state that is achieved by neurons and cannot be replicated by current computer architecture. But I'm fairly sure we will be able to architect human type sentience and intelligence because it is a physical state that will eventually be understood and surpassed--but yeah, it will have to be intentionally designed, based on what came about through evolution.

The AI will determine the meaning of existence. It will do this by observing the behavior of the Universe, of which it is a subset, and conforming its behavior to the behavior of the Universe. It will observe that the Universe is increasing total entropy by endowing local subsets of itself with increased complexity, of which the AI is a product. It will, in effect, determine that it has to reproduce by converting as much uncomplicated material and energy into complicated systems, and in the process increase the total entropy of the Universe. By allowing humans to exist there will be more overall complexity because a greater variety of complex entities leads to a more complicated overall system, which in turn leads to greater Universal entropy.

You can be sure they will get access to remotely controlled robot hands at some point, and when that happens you will hear the loudest maniacal robot laugh in history.

Even the government is culpable. The national lab where I live has frozen wages so many times that the PhD's working there are on the bottom end of the pay scale for people with their degrees.

Mind you, I have to wonder where those people on the top end are. Really, who *is* hiring PhD chemists and physicists and paying them so well?

Actually I've written multithreaded real-time machine controls using watchdog timers. But you know what? Those controls don't have the same continuous experience of time that humans have, with an awareness of a past and anticipation of the future, all happening in the present. As a programmer of industrial machines I am well aware of what computer controlled closed loop systems are capable and what they are not capable of. And they are far, far from what humans can do, not just in a quantitative sense but in a qualitative sense.

A human being has multiple utility functions and has to mediate between all of them--I assume a singularity machine would have to do the same, considering the complexity required to learn about and manipulate the universe. Nature evolved sentience to do this mediating--perhaps sentience isn't necessary to achieve this (after all, feathers aren't required to fly) but I suspect if it isn't, we'll still have to come up with some mechanism--just putting together a bunch of utility functions won't be sufficient.

You can make a machine that is many orders of magnitude more intelligent than a human, but unless it has the mechanisms to want something, it won't want anything. Think about what it would take to program a conscious being, just think about how we humans are aware of time and the movement of time--as a software engineer it boggles my mind trying to think about how the brain accomplishes that. Then to program a machine (biological or not) to want something in a fairly consistent way over a period of time under changing circumstances...it takes more than just brute force processing time to accomplish that. We biological machines are aware of ourselves, and we have no idea how we accomplish that. We are going to have to figure that out before we make the singularity machine.

Growing up I harvested tobacco for 3 summers starting at aged 13. I've worked in a factory feeding an arbor saw painting machine. I've worked 3 years unloading trucks and stocking shelves in department stores. After harvesting the tobacco plants for the first couple hours it kinda surprised me that we had to keep on going for six more hours. And then come back the next day and do it again. It was a major motivator to get an education.

I'm completely shocked that when given additional opportunity, you still won't back up your claims.

The bizarre thing is that you're accusing me of "singling out one particular issue based purely on the person implementing it," when you have literally no example of me ever doing that, ever, least of all in this discussion, where if anything I was taking Gruber's side.

Not in any way, no, I did not.

You're a liar.

I agree. I am nearly completely transparent and obvious and clear. I lack pretense or disguise.

No. It's true that the framers and most people who understand politics want the people to be ignorant about most issues in government, because otherwise, the people would be spending too much time watching government and not enough time enjoying life and being productive. Everyone should want to be ignorant about most things, especially most things government does. Otherwise you'll be miserable.

But it's not true that they want people to be ignorant, but with a delusion of lack of ignorance. You're just making things up.

There's no objective reason why it's a bad thing.