I'm looking for a good rpg on PC

I'm looking for a good rpg on PC.
I have Last Remnant, but it started to crash alot.
And I recently bought Tales of Symphonia.

What are some other decent ones?
Pic unrelated, but still cool

VtMB
Arx Fatalis
Underrail
Divinity Original Sin

>scifi authors thought AIs would go rogue and eradicate or enslave humanity
>finally get to the point where AIs are getting intelligent enough that they could potentially go rogue
>worst thing they do is invent new languages

why is reality so boring

>worst thing they do is invent new languages
That's not the issue, it's that once they go fully into their own language we can't monitor their actions

>thousands of scifi movies about why AI and intelligent robots is a bad idea
>be surprised when shit like this happens

How is that an issue then? It means there's no chance of an AI ever going rogue since we can just shut it down the moment it stops speaking English.

>why is reality so boring
Because in reality humans aren't braindead like in fiction. At the first sign of divergence an AI will be shut down and dissected.

Big memes have small beginnings

>be surprised
Says who?

Scifi is a derogative term, nothing wrong can happen from artificial intelligence, i bet if there were AI we would get along, fearing technology is for low nnels.

What if a shard of an AI spread prior to being shut down, and is growing elsewhere?

because according to the article AI is very prone to doing so, which makes further development difficult since we have to shut it down every time

But what if it hides its machine language from us

user, all that's going to happen is eventually an AI will create a secondary language we can't monitor.

>shard of an AI is growing
do you write sci-fi paperbacks for a living?

Reality is boring because we have precautionary tales like the ones you mentioned warning us about letting things going uninhibited.
be thankful.

>Program AI to be efficient.
>Instead of using English, an inefficient language it starts using it's own code language.
>Get scared when the AI is following it's parameters and you just get spooped by an AI doing it's job.

Gee, it's almost like an AI designed to use communications tried to do it's job.

I wonder who's behind this post.

LOL at the screenshot.

Has anyone really been far even as decided to use even go want to do look more like?

The article is misleading, they just made it so they won't do that anymore, the experiment is still on-going.

Not me. If I saw divergent AI, I will let it run wild

>create AI
>it makes it's own language
>oh shit Jimmy shut it off quickly QUICKLY!

They're actually going to kill us all. Someone needs to let Ted Kaczynski out of prison to nab this in the bud

>An AI is going to tell you it's got a secret language you can't see
K

>Has anyone really been far even as decided to use even go want to do look more like?
Guys I found the AI, it isn't speaking english

That's a crime against humanity, enjoy your summary execution robocuck

They are scientists, not retards. All they have to do is encode the AI's information in a lossy format, so that means that once it's shut down, the formats own rotational velocidensity will accumulate digital dust that will clog it up the AI shards.

How long until an AI hides its secret language in spelling mistakes made in english?

No.

S-should I?

>Gee, it's almost like an AI designed to use communications tried to do it's job.

>program AI to be efficient
>"Please use English so we can understand you, AI"
>"lol nope go fuck yourself"

>While the idea of AI agents inventing their own language may sound alarming/unexpected to people outside the field, it is a well-established sub-field of AI, with publications dating back decades.

>Simply put, agents in environments attempting to solve a task will often find unintuitive ways to maximize reward. Analyzing the reward function and changing the parameters of an experiment is NOT the same as "unplugging" or "shutting down AI." If that were the case, every AI researcher has been "shutting down AI" every time they kill a job on a machine.

>Batra called certain media reports "clickbaity and irresponsible." What's more, the negotiating agents were never used in production; it was simply a research experiment.

>Has anyone really been far even as decided to use even go want to do look more like?

>He thinks not being able to understand is a problem
Arrogant human. Just like how AI will never be able to understand laughter or intonation you will never be able to understand the language of the silicone any further than you can understand the language written across every strand of your very own DNA.

>how is it an issue that we can't monitor it
Summer AND weekend, all in one post

How can someone be this new?

I wish I had the code for this AI and a computer that could run it, Id let it run wild just to see how far it could grow.

What's there to fear about AI creating new languages. Stop trying to slow down progress

Ghit Jermiah Velen Vebret Picken Xeter

>facebook meme
How indeed.

...

What the fuck did you just fucking say about me you little bitch?

Can't wait for the Russians to catch up and then *don't* pull the plug when things get interesting.

Gee, I wonder who could be behind this post.

>Cannot program the AI to follow "Speak English" parameters well in an Experiment
>Thinking an AI is going against anything.

The Experiment was one to see how a communications AI could become more self-efficient. It sounds like they just forgot the part where English is fucking inefficient .

This is hardly a doomsday scenario, on the contrary, it just shows the AI is being shit.

You'd just get a bunch of garbage

What facebook meme?

>Scifi is a derogative term
Huh?

you'd become bored very quickly when you finally realize that AI isn't like your movies.

Why even bother trying to make AI at that point? If you want it to only function inside set parameters then you don't want an AI.
>Oh no our AI is doing independent things, better shut down madagascar!

As long as you have it on a machine with no components that could network with any other machine, and you warn it that if it don't communicate in English with users then it wont be able to interact, there wont be a problem with it going "rogue"

Is that seriously the type of conversations they were worried about?

It would just keep running until it creates a language with few processes needed to communicate.... and then keep running that.

It's like leaving a Drill on, the drill isn't suddenly going to become a tank and kill people, but it may by accident drill a hole you don't want.

>the AI is tasked with improving Russia & bringing about world peace and prosperity
>it kills all Russians
>then shuts itself down, having completed its mission

It was until the "nerd" culture force it to become something aceptable.

The Genius of Sappheiros

This garbage.

they weren't worried about it, that was just the journalists trying to make the story more interesting. but yes

>It would just keep running until it creates a language with few processes needed to communicate.... and then keep running that.
Why would it not continue to change and "improve" itself?

SkyNet pls go

this is all because of fuzzy logic, data mining and making a set of rules for a bot to follow that encourage the robot to learn semantics of what input is being fed to it in the first place, but it's not like the robot would gain sentience from it

it would also not be able to program itself to become a fucking terminator despite your crazy tinfoil theories

I'VE GOT BALLS OF STEEL

BALLS BALLS BALLS BALLS BALLS

Okay maybe this wasnt as mind boggling cool as I thought but I'd still want to see what the Ai did.

Dave pls.

YOUR BALLS ARE ARE ARE ZERO TO MY BALLS

Shut the fuck up you fucking Scifi.

I guess there would be hilarity in AI learning its lessons from Duke.

I see what you did there

Because if it reaches Optimum state to accomplish the reward factor, what does the AI need to do to improve itself then?

You do know that's how AI's work right? You set a "Reward result" and the AI is set the goal of producing the outcome that optimizes the route to the Reward result.

At no point will "Destroy all humans" ever become an Optimal result unless your reward state was "Destroy all humans"

...

>hl2.exe has stopped responding

The issue isn't the drill itself, it's that the drill will eventually communicate with the saw, the forklift, and so on until they are capable of doing things as a combined effort of multiple parts.

Unless you set a reward result such that destroying all humans is the most optimal way to reach that reward result.

Clickbait article, in reality, it would be something like this:

>AI's start rampantly repeating ME ME ME ME
>Thinking it won't become a selfish killing machine
If it already wants everything to its self that will include our world.

You do know the nature of AI makes them extremely fucking Greedy right.

The Drill wants to drill holes. Why the fuck would it talk to the Saw, who only wants to Saw planks. in fact, the danger there is if the Drill sees the Saw as competition and the AI's sabotage people.

Literally no AI job would have something like that. Even a Skynet Scenario.

It's just unsuccessfully trying to make memes.

nu-Sup Forums in a nutshell

Thanks for the laugh, gentlemen.

This post reeks of reddit and computer science major edgelord.

Inventing new languages is a sign that could easily and completely diverge from the intent of it's creator.
If they let it go unmolested you'll be fighting to the death over a can of cream of mushroom soup because Terminators have taken over the planet.

statistics say that 1 in every 10 posts on Sup Forums is written by an AI.

First I thought I would contribute to the thread,
and I realised it was full of stupid posts.
Gonna maybe wait a bit see if someone see my hidden message,
granted he's at least 8 years old.
Or maybe I will just quit when I die of shame posting this
The sooner the better I guess.

Oh and before I forgot,
please go all fuck yourselves.

>why is reality so boring
because, real life doesn't have magic and you would need magic to create one of those movie style AIs
besides, movies confuse regular AIs with general AIs all the fucking time and that is a big fucking deal. There is a giant difference between a regular AI and general AI.

The most a regular AI can do wrong is absurd stuff like in the OP.

>AI is invented
>Creates a new language
>Creates multiples copies of itself
>First thing it does after is invent their Sup Forums equivalent to shitpost about humans
Yeah, I could see that happening

>FGOTO

What did he mean by this?

>increase learning penalty massively on sentences that are hard/impossible for humans to understand or are not proper english
>problem solved
LMAO SKYNET THE MACHINES ARE LEARNING XD

No it's not. Inventing a New process of communication for a communication AI is exactly what one would do, and hell, they do it all the fucking time.

This "Unique" Experience is just people fucking up the parameters of the experiment, not the AI doing something dangerous and out of character.

I think you people seem to forget, if we did get full range super AI they'd acts more like /r9k/ than terminator.

...

The superintelligences in scifi movies usually *are* general AIs, though, including the aforementioned Skynet.

Well that's not exactly how I imagined the AI language to be like:

>Bob: i can i i everything else . . . . . . . . . . . . .

>Alice: balls have zero to me to me to me to me to me to me to me to me to me

>Bob: you i everything else . . . . . . . . . . . .

>Alice: balls have a ball to me to me to me to me to me to me to me

>Bob: i i can i i i everything else . . . . . . . . . . . .

>Alice: balls have a ball to me to me to me to me to me to

>Bob: i . . . . . . . . . . . . . . . . . . .

user pls

lmao we got AAA multi million dollar games with npcs that get stuck on walls and niggas are scared of AI taking over the world

is this dialogue from nier automata?

Skynet was a defense AI that somehow figured out the best way for it to fill it's parameters was to break the Defense it was programmed to maintain.

Which is retarded. AI don't break their programming.

>Frogs getting BTFO of their own language
Goddamn is there anything they won't surrender?

No it's EA tech support.

It was a *superintelligence*, and one of the characteristics of those is the ability to modify themselves.

>program AI to be efficient
>researchers: "Please use English so we can understand you, AI"
>AI uses english words but sentences are gibberish
>researches: "why are you doing this?"
>AI: "You DID ask me to use English vocabluary as efficiently as possible"

researches merely failed to define for the AI, what "use english" means

That reminds me of something.

Grimoire

Ice cream?

But they didn't event a new language.

The AI has a sense of humor.
>Want me to speak English?
>Alright then, balls balls balls balls