In one of my earlier posts, I posited that it’s difficult to test whether a simulated intelligence is self-aware or not (do we grant an AI rights and personhood). I concluded by asserting that choosing to believe in an AI’s self awareness is tantamount to choosing to believe in an invisible deity. This led to a series of conversations, some in the comments and more on Facebook, that got me to back away from this hard assertion.
Somewhat related to this, we now see discussions on whether we ourselves live in a simulated universe. Just last week in the Skeptics Guide to the Universe Episode 379 (starting around the 12:15 mark) they discussed that scientists have performed the small simulation of reality. The SGU then moved on to discuss a recently claim that we almost mathematically certainly exist in a simulated universe right now. (A description of this argument can be found here, although this is not necessarily the article to which the SGU was referring).
This got me to thinking.
Aside from the science-fiction story potential, the possibility that we are, ourselves, simulated beings is quite interesting. There are many implications to this if it turns out to be true. First, in considering this, I’m going to put aside any question about the soul, or faith. If you believe in a soul and your understanding is defined by religious teaching, this speculation is just silly, and there’s no point in engaging. However, if you contend that we don’t truly understand consciousness, and don’t automatically look to spiritual answers to explain it away, then this possibility bears thought.
First of all, if we live in a simulated universe, and we are also simulated intelligences experiencing self-awareness, this shoots to hell my earlier assertion on simulated intelligences being not self aware. If the math is correct that we almost certainly are simulated (or living in a simulated world), and given that I take my own self-awareness as axiomatic, then it is almost certain that I am simulated, and we, in turn, will create self-aware intelligences.
As I think through this, I realize I’m conflating two things. Living in a simulated world (like in the Matrix) is not the same as being a simulated being yourself, although the Matrix makes a case for the computer programs in the story also being sentient. So, let’s back away from ourselves being simulated for a bit and consider whether we live in a virtual reality (VR) or not. More specifically, let’s consider that we do live in such a world–what then would it mean? (Remember, sci-fi is speculative fiction… so as sci fi lovers, consumers, and writers, we should speculate on such things).
Folks who postulate along such lines also try to come up with ways on how we might test for such a thing. They’ve stated that a simulated world should have glitches, or discrete points of resolution (like pixels of reality). However, the SGU notes that we already now what the pixel of reality is–a planck length–and the pixel of time–a chronon— so why would someone capable of building such a perfect simulation not therefore model the world physics off of real world physics. In fact, this is exactly what is proposed… a simulation of the fundamental forces of the world should, over time, produce simulated life. In such a case, we have an untestable hypothesis.
This is what I refer to the “all-powerful deceiver hypothesis”. This is the idea that we live in a perfect simulation indistinguishable from reality. This also refers to the idea I’ve heard where young-Earth creationists told me that the Devil (or God) placed dinosaur bones and old geological evidence pointing to a long-earth history in order to test our belief in the Bible, in the idea that those who hold to faith (superstition in this case) even when presented with demonstrable facts get rewarded somehow. I find such ideas largely without merit… it’s not practical to treat this world as imaginary.
There is also the thought that if something it not provable or unprovable, then the burden of proof is on the one making the positive assertion. This is a claim made my atheists against the theist position–if you can neither prove or disprove God’s existence, then the burden of proof lies with the person making the extraordinary claim that there exists an all-powerful creator. Extraordinary claims require commensurate levels of extraordinary proof, and so the rational position becomes one of skepticism towards any god’s existence.
How is this different from the simulated world (SW) possibility? At least for the present, the claim is that probabilistic math backs up the SW hypothesis. Now, as I said, it’s not practical for me to go around informing my decisions over such a belief. In the absence of proof, I actively believe we live in a SW. But for the purposes of speculation…
…if we live in a SW, there is a creator, by definition. Not a god, but a person, people, or civilization that constructed the SW. Glitches could explain why people have non-repeatable experiences they interpret as miraculous. Maybe the gods are programs of some sort, but I would argue that such wouldn’t be necessary (if the SW is modeled off of the laws of physics).
…if we live in a SW, that implies there is a real world. There are still real physics, and non-simulated beings of intelligence and self-awareness.
…if we live in a SW, we don’t know how nested we are. It’s possible that we do experience an afterlife, at a higher SW level. If there is a RW at the end, there is still a finite point at which existence ends (without looking to transcendent or religious explanations).
But the idea that’s really neat to me… what if us figuring out we’re in a simulation is “the point”… it’s the test that proves to our creators (programmers?) that we (or the system, meaning our universe) is self-aware.
Ok, on that note, see y’all next week. And give the SGU a listen… really good discussion on this.