Imagine that, without any aliens having been ever found, we get to a point where we can build sub-atomically precise simulations of reality, spanning at least something the size (if not the likeness) of a solar system within the sim (if needed, we can scale up to galactic scales)... Assume we go about trying to create ('evolve' within the sim) our very own aliens, that is to say new organisms 'living' within these simulations, the likes of which have never existed on Earth.

Would we be able to create anything truly alien, or is our limited imagination and physical/biological convergence an inescapable boundary?

Good answers will have a rough sketch of how one would go about generating alien sim-life, or alternatively a decent explanation as to why it's not possible. As to what would qualify as 'truly alien', it's a hard call, but as a rule of the thumb, if I can imagine it relatively easily (say a griffon), it's probably not alien enough.

  • $\begingroup$ What's your thoughts on the relationship between this and the question of AGIs? It seems reasonable that, once you pass the singularity, AGIs have more imagination, by definition, so they could invent us such a alien race. If we have the computing power you mention, it seems reasonable that we may have AGIs. How does that fit into the direction you wanted to take the question? $\endgroup$
    – Cort Ammon
    Sep 9, 2015 at 15:28
  • 1
    $\begingroup$ We already have computer algorithms that work like evolution. Think about an alien planet with specific values (i. w. large swamps, no planet rotation,...). Then just theow the planet information into a coputer and let a evolutionary algirithm calculate generation after genation - always picking the one bacteria/creature that fits best into it's enviromemt. This takes lots of processing power but you'll have a very usable result $\endgroup$
    – BlueWizard
    Sep 15, 2015 at 16:17
  • $\begingroup$ If we had a computer capable of simulating a galaxy on the subatomic scale (and did so) I'm not really sure it would it be possible to not create alien species. No one is really sure how common abiogenesis is, but if what happened on earth isn't incredibly unlikely. You should be all but guaranteed to get alien life after some finite period of time. Whether or not you'd recognize it as alien enough is another question. It would probably be more different from people than, say, octopuses. $\endgroup$ Dec 27, 2016 at 4:27

2 Answers 2


I think we have demonstrated that our collective ability to imagine exceeds our physical form.

Consider the following creature:

It's a social creature with a very rigid role based hierarchy. It is a predator, and the sneakiest of predators. It sets a trap and quietly lies in wait for unsuspecting prey to go by. The predator identifies what the prey wants to see, and constructs an on-the-fly bait customized for this individual prey. If the prey sees the bait, and begins to fall for it, this predator carefully adjust itself so that it looks like there is nothing else in the world but the bait, and that no one else in the world could possibly get it. It's the perfect morsel for this prey. When the prey gets too close, SPLAT the prey is reduced to nothingness in moments by acid. Even if a prey escapes to tell the tale, nobody believes it because they never seem to be able to compare notes. After a failed attempt, it looks like no attack occurred at all. This is, in fact, why the creature is social. Success or failure, it rapidly disseminates knowledge of the result of the transaction to others of its cluster, all of which have specialized processing nodes which ensure any inconsistencies in the attack, which might help the prey outwit the predators, are smoothed over so that noone ever is the wiser.

Sound fantastic and potentially an exotic creature? I envisioned it right here and right now. It's also a SQL server. We deploy hundreds of them every day around the world. You can get a degree in managing them, if you want.

Chalk one up for imagination.

If a creature is advanced enough, its mind is going to be more important than its body structure. Thus the real challenge for making an "alien" alien is to create an alien feeling mind. To me, mathematics is in an excellent position to posit the ability to create such a mind. We have a long history of creating mathematical languages which can describe fantastical facts which had never been imagined before (complex numbers... seriously?). Thus I expect we would create a language describing fantastical creatures, and then use the simulation to build all of the "fine details" that end up actually creating the alien feeling mind.

  • 2
    $\begingroup$ I have never seen an SQL server spit acid. Not once. $\endgroup$ Mar 12, 2020 at 13:26
  • $\begingroup$ @user253751 I should really go back and make the connection to ACID properties more obvious. $\endgroup$
    – Cort Ammon
    Mar 12, 2020 at 15:17

If it were up to me, I would have the computer run many concurrent, smaller sims, modelling a few miles of randomly generated terrain and atmosphere (or lack thereof), with basic astronomy modelling (but not in detail). I would then try to randomly generate different kinds of cells and microorganisms. I assume this technology is within scope of a world that can design and run a solar system-scale simulation. Essentially I'd try to brute-force generate all possible microorganisms by sheer arrangement of atoms.

I would then have the computer process each micro-world as quickly as possible, modelling, who knows? ten thousand years per second perhaps, and simulate evolution. The evolution wouldn't necessarily be propelled by mutation and natural selection, but could follow all kinds of hard-to-predict paths. For all I know, it's reasonably likely that some of the life wouldn't experience aging, and some sort of immortal, self-improving life would emerge.

At any rate, at that point it would be purely up to physics, without human bias. Anything could happen. So since my system works as best as possible to brute force as many starting conditions as possible, based on chemistry which we can perform without significant human bias, and then relies on physics rather than imagination to develop the life forms, this method seems both reasonable considering the assumed level of technology available, and as free of human bias as we could hope to achieve.

  • 1
    $\begingroup$ This is the way I'd go. (It would also be a fantastic way of predicting which discovered extrasolar planets are likely to have life on them, which would in turn help direct our exploration of our interstellar neighbourhood.) $\endgroup$
    – Rain
    Nov 26, 2016 at 0:59

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .