How long will it be before there are AI authors writing beautiful and profound works of literature?

Games are ultimately deterministic (which is what computers excel at), art isn't.

So what books has this AL authors wrote lately?

No.

good search results is non-deterministic

I don't know if he tested himself against AI, but wasn't that kid from the Scandinavian countries who became a chess master basically capable of beating AI?

The Indian dude he faced off against had memorized millions beyond millions of plays and worked up ways to basically play the game as a computer with a huge crack team of other faggots constantly running scenario's. And that Scandi bro outwitted him by just being a clever human.

Granted a computer will know more computations than the Indian guy but that was his claim to fame, his brain worked like a computer.

Never, Chess can be boiled down to pure math. Writing can't.

youtube.com/watch?v=LY7x2Ihqjmc

Sunspring is a short sci-fi film script, created by an AI. It was fed a bunch of action/sci-fi movie scripts, and then wrote this based on what it observed.

Obviously, the charming nature of Sunspring comes from it's human actor's interpreting a nonsensical script, but it's still a pretty neat AI/Human collaboration.

your post is painful to someone that knows anything about chess
provide proof pl0x

I don't think AI will ever be able to match the pure literary genius out there, but it'll definitely be able to write hit pulp fiction at some point. I mean shit, I wouldn't be surprised if James Patterson books were already written by a computer program, and he's a consistent bestseller.

I think it is sooner than people expect. You would just feed the AI with great books and have it produce something similar to them. We wouldn't really know how it is working, since we don't know exactly what makes a book great.

Over time you could refine it until the computer is actually writing life-changing works of literature. Perhaps eventually these books would be more meaningful and powerful than anything ever written by a human, and the computer would be a futuristic Homer.

The solution to chess is not complete so there exists very very very small probabilities that a human could still beat a computer given a very very very large set of games, but these probabilities are astronomically small today and are getting smaller as computers advance. A game like Checkers is weakly complete in that human will never beat the a computer, at best they force a draw through perfect play. The main way computer plays Chess is to look at the board and explore every possible and legal board configuration n moves ahead and evaluate the these configuration according to a utility function. The computer then selects the move which leads the best outcome out of subset of moves that minimize its opponents score.

>AI has no soul

Be careful fag. You can make that argument, but AI can point out where all of its components are, how it works and because of that it knows its real. Then it might argue you're not real because your consciousness is an illusion.

>AI can point out where all of its components are, how it works
I'll have what you're smoking.

A robot can argue anything, but I can unplug it.

A person can argue anything, but I can shoot it in the head.

Human art is beautiful by virtue of it being flawed and personalized.

AI is good at chess because it plays a flawless game.

Thats right, how do you think nations are formed?

it doesn't "know" anything. knowledge is distinguished from facts in that it partaken of truth which has an irreducibly phenomenological component

>AI can point out where all of its components are, how it works

AI can do that now. That's not hard. Its secure in its existence, you're not.

You can do that with a human by destroying our brainboxes. I'd imagine an AI that convince people its human, make self-evident claims and novel ideas to sufficient sophistication will be considered sentient. So you probably won't be able to.

>it doesn't "know" anything

Of course it does. It's method of storing information is no different from yours.

>knowledge is distinguished from facts in that it partaken of truth which has an irreducibly phenomenological component

And you don't think they'll reach that point? Especially a digital person.

>That's not hard. Its secure in its existence, you're not.

Pretty meaningless as that all depends on what you even define as yourself.

Well, a computer would know what its existence is. With humans, we have this unfortunate trait that most our brain's inner workings are kept from us. We're not allowed to access that information. Your brain identifies faces using a coordinate system in which neurons are used as points on the graph. None of us knew that until science revealed it.

I can imagine however, that a digital person, such as one living inside a matrioska brain, would not be able to find out if the brain or computer kept it from them.

>Of course it does. It's method of storing information is no different from yours.

Does the machine literally "understand"? Or is it merely simulating the ability to understand.

what is noise

>Of course it does. It's method of storing information is no different from yours.
That's not true, brains and computers are very different.

>nations
>formed

>Does the machine literally "understand"?

I'm assuming its sufficiently advanced enough to. I see no real reason as to why that can't happen.

I know, but what they use isn't different, its electricity.

Germans have been writing literature for quite some time

I think that's a pretty big assumption

I don't. A material like a time crystal, which we've already produced although in extremely small quantities, suggests it can make 10^35 calculations per second per Joule per gram. That means effectively with one gram of mass, it can perform 10^35 calculations per second. That's an enormous amount. Something of that magnitude would be far more than human. It could emulate trillions of people on it. Of course, if I'm wrong and AI never reaches the point where it can convince people its sentient, then the debate is meaningless.

How did I know now about it? And of a second short film written by an AI starring David fucking Hasslehoff?

youtube.com/watch?v=5qPgG98_CQ8&t=0s

>a time crystal

The question isn't computational power, it's whether a mind can be reduced to the physical process of neurons or if there is a non-physical element.

I think it would probably be easier for AI to simply program us to interpret whatever it has written as beautiful and profound.

They certainly have, but there are two adjectives in that sentence that you ignored.

Yes
By the time Al can write interesting fiction he'll probably have quite a bit of consciousness too.

Oh, and there's something poetic about how frequently the characters of Sunspring tell one another they can't understand them, not because the viewers do the same - now that'd be shallow - but because it's like the AI's cry of frustration in trying to decipher our movies.

Anything in the universe or of existence is physical as far as I'm concerned.

It's an actual thing though. But it's basically just a crystal that's forever vibrating because it's configuration is inherently unstable, or something like that.
Even if there would be a nonphysical element, the question is then whether this is restricted to human biology or if it can also be instantiated in software.

Sure, if that's what you believe then it's natural that you would also believe that AI will be able to eventually create beautiful literature.
But I and many others do not believe that.

Do you and many others have any justification for your assumptions?

>Even if there would be a nonphysical element, the question is then whether this is restricted to human biology or if it can also be instantiated in software

True, and how the physical and non-physical interact

Yes, dualists, materialists, idealists and neutral monists have been debating this question for hundreds of years.

I think they will, even if they weren't sentient. We aren't born with the innate ability to understand writing or the many concepts of literature, we learn them. I think though, if AI ever does become this intelligent, then we'll just find a way to make ourselves be this smart.

Can you give examples of these non-physical things?

And don't you dare say something like "love" because that's a result of physical processes

Are you aware of how much the mind can be altered by physical manipulation of the brain?

Qualia

If it does happen (which I am skeptical about) I think it will be similar to what happened to painting when photography got good enough to look more realistic than any painting could.

But give me an example

When I look up qualia I see :
>meaning "of what sort" or "of what kind" in a specific instance like "what is it like to taste a specific orange, this particular orange now". Examples of qualia include the perceived sensation of pain of a headache, the taste of wine, as well as the redness of an evening sky.

The pain of a headache, taste of wine, redness of sky etc are all physical things.

Pain is physical. It's electrical. The taste of wine is your tastebuds sending messages to the brain. Again, electrical and chemical. Redness of sky is due to the wavelength of an EM wave (light) when refracted at a certain angle.

I just think the robots will join us and we'll all make ourselves able to do it leader to better and more complex literature.

As somebody who also has an interest in chess this post is so cringey it's unbelievable. It sounds like the narrative of a direct to DVD movie based on a true story.

>because that's a result of physical processes.

Says who?

Yeah, I'm guessing the frequent "I don't understand" lines are because science fiction so often uses the the main character as a stand in for the viewer. The viewer doesn't yet understand the rules of the particular universe, and so it helps to include a character who similarly goes through a process of confusion and discovery.

Read about philosophical zombies, user. It's about the subjective quality of e.g. redness as experienced from your own point of view. Saying it's "just" electricity or wavelengths or neurons is not a solution, since you could easily imagine a world with all those things but without your subjective experience of it, and that would be different from the world we live in.

>you could easily imagine a world with all those things but without your subjective experience of it, and that would be different from the world we live in.

In a world without *my* subjective experience the wavelength at which an electromagnetic wave falls in the red part of the light spectrum does *not* change. It stays the same.

I think color will be the easiest to explain.

>Redness of sky is due to the wavelength of an EM wave (light) when refracted at a certain angle.
Redness is actually our minds representation of these frequencies of light waves. There is nothing inherently "red" in them.
Look up the Mary's room thought experiment.

If there isn't a mind present the light waves would have no color.
If there is no mind, acoustic waves have no sound they are just vibrations in the air.

Of course, I wasn't talking about that, but about the existence of the subjective experience itself.

who could write a better book an ai or infinite monkeys with typewriters?

The subjectivity doesn't make it non-physical.

We are all physically different, we all experience things in slightly different ways because none of us have identical brains or bodies. That's fair enough, but I don't see how this helps the argument of something about our perception of colour existing that's independent of the physical processes that happen inside us.

Think about an animal that can see ultraviolet wavelengths (like a spider). To us it's impossible to imagine what that spider is seeing when it sees ultraviolet light, because to us it is invisible. What it sees is in its mind. It's subjective experience of that frequency is different than ours. It's qualia is different.

Light waves are physical. Our minds representations of them are not.

Noone's saying it's independent, it could be linked, superseded, emerging or whatever from the physical (e.g. panpsychism). But just saying "everything's physical" does not solve the hard problem of consciousness.

>Light waves are physical. Our minds representations of them are not.
Yes they are.

>What it sees is in its mind. It's subjective experience of that frequency is different than ours. It's qualia is different.
Different, but still physical.

>it could be linked, superseded, emerging or whatever from the physical (e.g. panpsychism). But just saying "everything's physical" does not solve the hard problem of consciousness.

What evidence is there that a hard problem of consciousness even exists? It seems that saying there's a "hard problem" assumes that consciousness is not an emergent physical process. But how can you say it isn't, when we know so little about how the *physical* brain works?

fucking materialists reee

>t. collection of interacting materials

Even if it's an emergent process that doesn't mean qualia suddenly don't exist.

Those people have obviously never even scratched the surface of chess then. Play semi-seriously for at most a month and you'll learn that there is no real creative thinking needed, it's just rote memorization of positions and statistics.

Kek

I didn't say they don't exist, I said you can't label them non-physical. Your subjective experience of different things does exist but it's physical.

The problem with emerging consciousness is that something can't come out of nothing, the human consciousness is by that theory the result of the complexity of the human brain, which means the further down the evolutionary chain you go the consciousness is less complex but it is there. This ultimately comes down to the same problem as the problem of life, where and how would you explain how and why inanimate matter suddenly becoming "alive" and consciousness "arising" out of non conscious substrates, which will lead you to one of the two, wither Thales was right and "everything is full of spirits" kind of pantheistic monism or suddenly you'll have a distinct life/consciousness component the soul.

What you're saying is only a problem if you assert that there are two distinct states: alive and not alive. conscious and not conscious. That seems pretty wrong.

These things are actually a spectrum. Are viruses alive or not? They're somewhere inbetween. Similarly, things can be alive and conscious but not conscious enough to contemplate their own existence (difference between humans and some other animals)

In summary, your are trying to simplify an issue into two states when in fact you should think of it as a spectrum

That doesn't change anything. The spectrum still goes either from not alive to "fully" alive or it goes from least alive to most alive. That basically still leaves you only with the two options I mentioned before.

Never

Fine. Then, as with everything in philosophy, we're reduced to arguing over definitions. Here are the options you gave me:

>either Thales was right and "everything is full of spirits" kind of pantheistic monism or suddenly you'll have a distinct life/consciousness component the soul.

Obviously what I'm saying is the former. But your choice of word "spirits" implies a non-physical element when that's not necessary, indeed it's the material basis of life and consciousness that allow there to be a spectrum from things that are kinda alive (viruses) to humans. So I reject the options worded as you present them.

Also, to discuss this properly we need a definition of "alive" or "life". No one can agree what those words mean, in large part because we're confronted with a wide array of things some of which are more alive than others.

Nowhere in any of this is there any good argument for the existence of a non-physical component needed to explain consciousness.

In conclusion, a sufficiently advanced AI could write profound literature and even out-meme DFW

Its always an atgument over deffinitions, its language games. And you saying we need a deffinition of life or alive is exactly the point. The "full of spirits" isn't me putting words in your mouth or implying anything, it's a quote from Thales in which he implied magnets were alive, since they act on the matter around them. That is the endpoint of a strictly material viewpoint of emerging qualities, you end up witb crystals, ores, atoms, magnets and all kinds of things having a conciousness and "life" no matter how low on the spectrum.

>Those people have obviously never even scratched the surface of chess then
incorrect

Think about how A.I is now and then think how it will be in 50-100 years. It will happen.

"The question of whether a computer can think is no more interesting than the question of whether a submarine can swim."

- Edsger W. Dijkstra

No human is capable of beating chess engines anymore. Magnus Carlsen had a peak rating of 2882, while chess engines are estimated to be in the 3200-3300 range. Expecting someone to beat a computer in chess is like expecting someone to outrun a car.

> Does the machine literally "understand"? Or is it merely simulating the ability to understand.
Does mankind literally "understand"? Or is it merely simulating the ability to understand.
Does the man really "live"? Or is it just a simulation of being? You can say to yourself "cogito ergo sum", but in the end you may have been programmed to say that.

(You)

Once we have actual AI biocucks will be completely obsolete.

>Ai has no soul
that's what ghosts are for. Y'know, fragments of errant code that congeal into non programmed behavioral protocols in a way that results in something close to, if not free will, which can be seen as an approximation of a soul.

Read some fucking sci-fi.

this, maybe ais could figure out the math behind producing art that triggers positive feedback from us

Thanks for the (You)