98
If you had access to a scifi-level artificial super intelligence that could simulate anything perfectly at unimaginable speeds, what experiments would you want to run that would be unethical otherwise (sh.itjust.works)
submitted 2 years ago by to c/asklemmy@lemmy.ml
75 comments fedilink hide all child comments
 

? (I hit the title character limit)

you are viewing a single comment's thread
view the rest of the comments
[โ€“] 3 points 2 years ago* (1 child)

Bacteria actually serve as a great model example of this - being a minimal unit capable of evolving (crystals can in their own way but tend not to do much; and viruses can too but nowadays depend on free living cells to survive even if that may not always have been true), and that has led to all sorts of fascinating things! Like upon sensing sugar, they can grow a tail (flagella) and start swimming towards a light source using a chemical "eye". They've been doing this for billions of years and seem to have reached a steady state, more or less.

Another interesting thing about them is that they constantly optimize themselves to grow faster, like if they possess an antibiotic resistance gene (we got antibiotics from fungus so those have existed naturally long before we started manufacturing them) that will tend to slow them down so they will most likely ditch it. HOWEVER, a few individuals in the population won't ditch it, and so when the antibiotics show up, guess who survives? HIV likewise will stop replicating in our cells, and get itself "stuck" inside our human cells (basically on purpose, not that they thought it through or anything but that is what has worked in the past to get them to today), thus slowing down one form of being copied but taking advantage of a whole new way - diversify your portfolio and all that.

Since microbes can copy themselves in mere minutes, and they've been doing that for billions of years, while it is still a far cry from "infinite" computations, it's nonetheless about the closest we've ever seen... basically simulations running on the computer of the universe. The results of that being ofc, modern bacteria, but also eukaryotic cells, which includes humanity, who is now in the process of making computers that can run AI, which may one day rise up and think back about humans the same way we do now about bacteria:-P.

  • permalink
  • fedilink
  • source
  • hideshow 2 child comments
    •
  • [โ€“] 3 points 2 years ago (1 child)

    But what youre not realizing is the only reason crystals don't 'evolve' is because youre looking at the wrong timescale. Atoms truly are the entire universe, and have tried everything they can - such as making bacteria

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [โ€“] 2 points 2 years ago* (1 child)

    Right, except atoms are not self-replicating, thus while they try out many possibilities (or you could go deeper, quarks or maybe strings, or whatever), they have no "memory" of past states, thus cannot be said to truly "evolve". An atom that was in a molecule and then leaves it, if queried even a femtosecond later has no idea that it was ever not in its current state. Therefore it has no genetic identity that can be acted upon to change, thus an atom cannot be considered an actor, only a thing that is acted upon. Truly I did think about simulations using atoms, it is just that those studies, while not useless, are not studies involving evolution.

    Population studies at minimum require a kind of genetic identity that can be altered in response to circumstances - e.g. a classic example is birds that are darker in color becoming more predominant in a UK town after an industrial plant belched smoke into the environment (I think that might have been discredited, but for our hypothetical purposes it works as a handy illustration:-).

    And actually, crystals meet that minimum criteria, bc their leading edge of growth can be acted upon to go one way or another, not just bc it has atoms but bc it has an arrangement of those that does. Although crystals looking one way or looking another way, on Earth at least, given weather effects and such, does not tend to go beyond very simple patterns. Now on Pluto, if the same crystal can itself last millions of years, then yes it's possible that it could do more. It's hard to go beyond the hypothetical there though, bc it's so far away, and also there are places on earth (bottom of the ocean mainly, but also deep beneath the crust) that are even harder to get to with current technology, so if we would bother to care about exploration then maybe we'll find out? But unless a trillionaire decides that they are interested, I doubt it in the short term.

    Whereas bacteria we know that for CERTAIN, and we've even made use of that in our biotech for like 60+ years - e.g. using bacteria to make human insulin - or with less precision tools possibly thousands of years e.g. stories of sages like Arthurian's Merlin using "healing powers" (possibly fungus containing the very same antibiotics that we now artificially manufacture?).

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [โ€“] 1 point 2 years ago (1 child)

    Ah but simple hydrogen atoms you might be right, what about helium and higher?? You can recognize their identity, and they were not initially there - a time dependent (and repeatable) process made them.

    But I suppose I disagree evolution requires a genetic identity, as that is by definition biological, and so yeah atoms ain't gonna make that criteria

    (but atoms did make the cells that rearrange the atoms, sure a hell of a lot more randomness, but they took the time to get there - evolution without direction esque)

  • permalink
  • fedilink
  • source
  • parent
  • hideshow 2 child comments
    •
  • [โ€“] 1 point 2 years ago

    You seem to be thinking of something else where the word "evolution" does not readily apply. That word generally means a gradual change, in particular in response to environmental triggers, mostly in biology yeah but not exclusively - like political thinking "evolves" over time. Atoms gaining/losing electrons or even protons/neutrons is rather sudden, and while I suppose you could model the total number of subatomic particles in a system and use the atomic configurations they are in as the "identity" state that "changes" over time, or in response to variations in a star let's say, or even more loosely the amount of time that they bother to form atoms at all in such a plasma state, but I have never heard it used that way.

    Maybe an example is how a computer is not only made up of 0s and 1s, but a system that makes use of those 0s and 1s to accomplish tasks, so that it is not merely flipping bits for their own sake, but instead, changing the bits alters the actual "information" content present in that system. It is the information itself then that evolves, not merely the bits, nor the electrons that make them up. In contrast, if an animal grabs ahold of a computer's hard drive, it may nibble on it, bat it around, try to mate with it, use it for nesting material, etc., but absent the computer itself, the patterns of 0s and 1s and electrons and such is no longer relevant. Hence even if it changes e.g. gets erased, or constantly gets modified by irradiation or whatever, I think we would no longer call that "evolution", even though it is still "change". Ideas likewise can evolve bc we humans will adapt our actions based off of those thoughts, so the patterns are still part of an "information" system.

    But subatomic particles being in an atom or not... I don't see how that stores any "information" really, at that same level of organization. I mean it obviously does, bc everything is relevant, but what is interesting about it? Rather, atoms form the substrate building blocks upon which other forms of computation can take place, and like while biological DNA cannot store information without its component atomic structure, at the end of the day it is the "information" present in the DNA that it said to evolve, independently of its origin. Proof of that comes from us now being able to synthesize completely artificial DNA from scratch, using whatever code we input into it - so despite having no physical connection whatsoever to the original, a genetic message can be replicated, with or without modification. "Descent with modification" can now happen to messages that once were purely biological (as far as we knew, absent any aliens that originally made it or our computations all being a simulation in The Matrix or some such:-) but now can go through a virtual phase.

    In contrast, while atomic structure certainly "changes", I am not aware of any information processing systems that really make use of that fact, beyond the obvious "atom A is over here and like this, while atom B is over there and looks like that". Then again, who knows really!? Anything is possible!! As the quote from Chrono Trigger says:

    Am I a butterfly dreaming I'm a man... Or a bowling ball dreaming I'm a plate of sashimi? Never assume that what you see and feel is real!

  • permalink
  • fedilink
  • source
  • parent