@Cirrus Light  
Just because you feel like you experience something doesn’t necessarily means that it actually exists as a real concept. The mere fact that we feel like we have free will isn’t enough to prove that we have it.
 
Also, quantum mechanics being probabalistic doesn’t lead to anything like free will, because it’s hard coded into the theory that nothing but the physical laws therein influence said probabilities. There is no way to get quantum probabilities to lead to true free will, as even if quantum effects routinely affect the brain’s operations on a macro scale (which is incredibly unlikely, both because neurons do not transmit information using individual fundamental particles, and because singular neurons cannot notably affect the brain on their own), said changes would not be within our control. As a physicist, I can tell you that you’re going to have to look elsewhere to find a source of free will.
 
Personally, I don’t believe that qualia can actually exist, and that if we had sufficient knowledge of the brain, we could exactly measure states of excitement, joy etc. The brain isn’t a magical object that can rise above the laws of physics despite being made of physical components and operating according to said laws.

@Eeveeinheat  
I take free will to be on the same level of qualia - something you experience and thus know exists, but isn’t physically provable (for that matter, every unique experience of qualia is that way - the feeling of pain, love, joy, confusion, surprise, excitement, etc. etc. etc.). That’s why quantum probabilities strike me as such a nice possibility. It’s a way that existential/“solipsistic” reality can meet physical reality.
 
I put solipsistic in quotation marks because I’m pretty much solipsistic, except that I assume that somehow other qualia interact through the physical universe. It’s an assumption as well as a religious belief. So I reject the solipsistic notion that I’m the only qualia in the universe, but accept the rest of it - that physical reality isn’t fundamentally different from dreamscapes. It’s different enough that I’m far more invested in it, but not fundamentally a different thing, in some ways.
 

 
@Fwelin  
There’s no distinction anywhere, actually. It’s just that at scales of nanometers and larger, quantum probability covers such a relatively small area as to be negligable or so unlikely that you can safely ignore that likelihood. The universe, however, continues to decohere at all scales and in all ways in Many Worlds interpretation, which is the only interpretation that really makes sense to carry to scales outside the quantum.
 
The fascinating thing about the brain, though, is that it’s an immensely chaotic system, where perhaps(?) a few quantum state changes could lead to a much larger cascade of differences. As an example, imagine a series of dominoes lined up, with each one slightly bigger, where a quantum probabilistic event can have results that affect macroscopic scales.
 
In actuality, this would equate to a synapse not having quite enough energy to make the jump, but a quantum state (or a few thousand?) giving it the extra bit it needs to make the jump, resulting in a cascade of activity throughout the brain.

@Cirrus Light  
Yeah. The turing test is bullshit.
 
@Fwelin  
There isn’t one because no one can agree on a definition of free will. They all think they know what it is but when they try to explain it, it never makes any logical sense. “free will means I can make my own decisions!” or “free will means nothing but me is in control of me!” “free will means if the universe were copied a few times, then each of my copies could make their own decisions that are different!” are some examples. I for one find the question has already be answered, as it’s objectively proven that your consciousness (not qualia, though) is 100% a result of field interactions. Which is a fancy way of saying ‘physical’. External factors and internal facts out of your control change not only how you behave, but also how you literally think.
 
But I don’t care that there isn’t free will. I feel like I have free will, so it’s all the same to me. What I want to know is what my qualia is, not if my consciousness is free or not. I will always be my qualia, but my conscious mind isn’t constant. So why put that in high regard?

@Cirrus Light  
But then, what exactly is the physical source of the divide between “probabalistic” and “deterministic”, in terms of theoretical free will?

@Badumsquish  
Why not straight-up make a pillow pony? Maybe even make it afraid of the dark, so a cover isn’t needed, but still have it absolutely love cuddling? I dunno… when I thought of “pillow pony”, my mind just started going every single direction with possibilities.

@Eeveeinheat  
A perfectly reasonable view - one I’ve kind of accepted, myself. I think I didn’t mention it because it’s fun to see what people come up with - though you and I in particular have had this conversation before, I believe.
 
 
@Eeveeinheat  
A very nice qualification to add - If you can objectively prove it only follows programming in a purely deterministic way…
 
But then again, while I think we’re endowed free will through probability, there’s a chance that thought is wrong, in which case they could be sentient even if they are purely deterministic.
 
Though if deterministic AI’s could be sentient, then sentience would also apply to anything at all - seesaws, dakis irl, your computer - and here’s where the line would blur - dating sim characters, basic AI’s that already exist in games, intermediate AIs like Siri and more advanced ones…
 

 
 
In any case, while I agree with the “treat it sentient because we can’t tell”, I remain curious about their sentience. Wiki or Google the “Chinese Room Thought Experiment” for a disproof of the turing test.
 
Eeveeinheat in particular is probably aware of this, but anyone else following the convo probably isn’t.

@LogicalExtropolation  
I follow the mechanics and the turing test. If it thinks like a human and acts like a human, then you treat it like a human. If it acts like a human, but you can objectively prove it is not thinking like a human because it physically cannot, then you don’t need to give it rights. The type of AI I am talking about is one that thinks and acts like a human but you are dubious if it has qualia like a human – an untestible quantity.
 
@Badumsquish  
Agreed.

@Badumsquish  
Most of those have actual reasons behind why they’re considered good or bad, though, even if they’re sometimes weak.
 
Regardless, cultural conditioning is a far cry from actual brainwashing or mental programming.

@Eeveeinheat  
I don’t know about how “moral” it is, but… that’s kind of how we live in my experience. Why’s nudity bad? Because it offends people. Why does it offend people? Because we’re told from birth it does. Why are we told from birth that it does? Because that’s just how it’s always been :P
 
Same could be said for dirty words(fuck is a swear but frick isn’t despite having identical meanings and usage), stereotypes (Black man dressed like a gangbanger: OFFENSIVE STEREOTYPE. Australian dressed like Crocodile Dundee: funny). We’re literally programmed to think these things are good or bad from day one.
 
@PegaSUS  
Nope, the pillow is a part of her. It’s alive just like she is :P

If only she wasn’t “Safe”… we might have a conversation with her and the explicit OC of lil miss jay, “Pillow Talk”.
 
… Not even sure how that would go. Probably the anthro-pony falling asleep while the pillow tries talking to her before realising she’s being ignored.

@Eeveeinheat  
I always follow the Turing Test. If during a conversation with an entity you cannot tell that it is not human, then you must there after treat it as a human, because there is no identifiable difference between it and a human acting like it, nor vice versa. Mostly because I’ve had very vivid nightmares about getting stuck as some whatsit and desperately failing to convince anyone that I deserve to live, let alone be treated as the person I clearly am.
 
That’s also why I used to get really angry about Marche in FF Tactics Advance. I thought of myself as one of the inhabitants of the world he said wasn’t real, one of the people he fought and struggled beside on his journey to unmake everything they’d ever known, and I fantasized about having the others hold him down, prying open his eye, cutting my palm, and bleeding into his eye until he told me I was real.

@Cirrus Light  

 
It doesn’t matter. If you can’t tell if it actually has qualia or not, and the machanics are the same as how the brain works, then you must assume it has a qualia. To do otherwise would be amoral. You do this with other humans, after all.

@Eeveeinheat  
I guess I should rephrase this; the issue isn’t if we can ever build one, the real issue is, if we build one, how can we tell it apart from a philosophical zombie?
 
I conjecture there’s something to do with quantum state collapse/Everett Decoherence that’s related to free will - after all, quantum state collapse is random and Decoherence - in that model of quantum mechanics, everything happens (the particle goes through both slits - but in different parallel universes), but obviously your qualia only follows one “worldline”, so to speak - so how does it “choose” which world to go into at each decoherence event? It seems like a natural and fitting place for free will to exist.
 
To me, this would suggest that if there’s a quantum uncertainty somewhere, then it truly has “free will”.
 
But I can’t prove it, I can only provide the argument that it seems natural and fitting. I may be wrong. There’s really no way to know for sure, thank to the Mind-Body problem. We’re always stuck not being able to prove that our creation isn’t a philosophical zombie.
 

 
…This Daki pone is very much like the paintings in Harry Potter in that regard…

@Cirrus Light  
[quote]Anyways, this gets into some even deeper questions. Can you ever create a sentient AI, or will it always just be an increasingly complex philosophical zombie – something that acts like it feels because it’s been programmed to, but doesn’t actually feel, any more than a steel beam feels when it’s kicked?[/quote]  
Yes. 100% yes. No doubt that we can create a sentient AI. The laws of physics allows us to feel, so it can be replicated. Even if it ends up being partlly based, or fully, on organics, it would still be artifical if it was designed by humans. This isn’t even a question of if – it’s a question of when and how.
 
I don’t think it will be possible to build a sentient AI that truly feels using logic gates and binary code but there is no doubt in my mind that we will discover the best way to create a neural network equal or superior to our own brains one day. We are hardly the pinnacle of evolution, after all. We are just “good enough” to breed.

@Eeveeinheat  
The difference is that humans are biological. While in a fundamental sense, that’s unimportant, in this particular application I think it is, and here’s why;
 
We don’t know what actually makes a human “happy”, content, the most meaningful existence, or even which one of those are the most important. At best, we have a nebulous understanding of what humans, and our perceptions are. Can you actually “program” a human being to be happy to be abused? Maybe, but it would certainly make them very different from what humans are “supposed” to be like.
 
It’s scary territory to enter, but I think, based on our western values, we can establish a few things that seem intrinsically moral to us. It’s not moral, for example, to force someone into subjugation. It’s questionable to take them willingly - and it’s considered rape if you alter their mind so they want that, and I don’t see why this would be any different - a contract signed under coercion, for example, is not considered legally binding.
 
We value human free will. Our right and ability to make decisions, and to make us who we are. This is a very good western value.
 
But so far I’ve only been talking about humans. What about artificial constructs?
 
Well, to start, is porn immoral for what it does - how it treats drawings? What about grimdark pictures of the mane six? Well, no, they’re just drawings. What about “Banned from Equestria Daily”? You interact with those characters, and they’ve been programmed as though they want certain things and have certain motives. Is it immoral to give those characters in that game such programming?
 
I think not. Any construct is nothing more than an increasingly complex version of, say, a character from a dating sim.
 
You created that character from the ground up. You manually built in every little bit of them and who they are. I don’t think there’s anything immoral about giving them a clear purpose in life - like a cutie mark - when that’s just part of programming them. The alternative would be to make them either feel no purpose in life (deep depression) or not know what their purpose is, when in fact, they do have a very express and distinct purpose. Doing the latter will do nothing but cause unnecessary pain for everyone involved. Giving them a purpose is mutually beneficial to everyone.
 
The difference is that they don’t have that free will that we do. We can trace everything that happens in them back to simple, deterministic mechanical actions. They, quite demonstratably, have no free will. We built them from the ground up, every little bit.
 
A human being on the other hand, is grown, and builds itself from the raw materials provided to the mother and themself.
 
If you made the AI a “learning” AI - that is, it wrote its own programming as it grew up, then I think it would be immoral to later hamfist its head against its will.
 
I’m not sure why, but that feels right to me - I think it has something to do with randomness and free will.
 

 
Anyways, this gets into some even deeper questions. Can you ever create a sentient AI, or will it always just be an increasingly complex philosophical zombie - something that acts like it feels because it’s been programmed to, but doesn’t actually feel, any more than a steel beam feels when it’s kicked?
 

 
@PegaSUS  
I’ll bet getting washed is really fun since she doesn’t have to breathe :q

@Badumsquish  
Is that so?
 
Now, is she to forever stick to that specific pillow (case) or could could she move through adjacent ones?
 
@Eeveeinheat  
Yikes, that’s deep.

@Parallel Black  
That’s why I think there are two major paths for AI. Free AI, or programmed AI. Most people will justify thr programmed AI by saying they are happy, so it’s no big deal. Kind of like how some people justify how, ahem, Muslim women in certain countries cover their faces in public. “It’s their culture. It’s how they were raised. It’s all they know, and they seem fine with it!”
 
Okay, this example is getting out of hand and turning political. But I think I made my point. What constitutes being free? What is the difference bewteen a human programmed from birth and an AI programmed from birth besides the time it takes to program? This is a moral quandary most people don’t ask, and I bet most people find irrelevant. But I feel it’s crucial to understanding how we view free will.

@Eeveeinheat  
I would say it’s horribly immoral but unavoidable if that AI/magically imbued Golem is going to be anything more complex than a machine. On one hand a simple machine can’t learn and grow over time, while on the other, a truly free Golem would always end up wanting to try new things and in turn, stop performing their intended purpose.

@Badumsquish  
That works :D
 
@Eeveeinheat  
._.

@Eeveeinheat  
…Duuuuuuuuuuuuuude!

@Badumsquish  
“Definitely inspired by it. I like the idea that such creatures can just be manufactured and purchased, yet happy and fulfilled in their existence.”  
If you think about it, this idea is really freaking dark. Especially because AI will become a reality and this is the kind of thing we will do to get them to serve us. They will feel enjoyment in whatever they do because we programmed that. They will feel fulfilled in whatever they do because they were programmed that way. They will feel depression if they don’t do what they were meant to do because we will program that.
 
Is this moral? I don’t know. They are happy. They would enjoy their life. But it would be little more than slavery via mental programming. Is it wrong to bring something into this world just to use it? You could argue that’s what parents do with their biological kids already. They want them because they want them. Not because they want to bring a life into this world just so the life can exist but because it feels good to want a kid. And it feels good to mentally program the kid to your specifications (teaching).
 
So, would you say this is moral or not?

@Dirty Bit  
Mine doesn’t even come close to that. That was super adorbs and powerful :D
 
@SuperSupermario24  
D:  
Just ask and she’ll read a book to you while you drift off instead :D
 
@Chicago Ted  
Definitely inspired by it. I like the idea that such creatures can just be manufactured and purchased, yet happy and fulfilled in their existence.
 
@FanOfMostEverything  
@AwkwardLuna  
She’s a living, thinking, freely-moving image of a pony on a pillow that can freely move her pillow around, and is completely aware of her existence as an object of comfort but also completely fulfilled and content in her existence.
 
@PegaSUS  
Her image can move freely about the pillow :D
 
@Zebasiz  
She can also cuddle you while you rest your head on her while she speaks :D

Aaaand then you stic-… SAFE!?
 
… Well… so it is… very well… she is cute.

…I would actually really like a pillow like this…

This needs the Badumsquish Strikes Again tag, because my sides sure felt it.