Will GPT4 be able to Mod in Games?

[BernL]

Will GPT4 be able to Mod in Games? you know, better programs, creative thinking to create Mods? Could it re write RF2 And add features??? Or ask it to write a program that allows say RF2 to use more cores? I mean what could one prompt it to do?

 

[demetri]

Since the current AI demonstrates lack of good old logic and zero basic math skills I highly doubt it can actually program anything remotely complex. Good luck asking it to write a multi-threaded RF2

 

[Deleted member 197115]

Just for fun I asked Google Bard.


It is possible that GPT-4 could be used to mod games in the future. GPT-4 is a large language model that has been trained on a massive dataset of text and code. This means that it has a deep understanding of language and how to use it to generate text, translate languages, write different kinds of creative content, and answer questions in an informative way.

If GPT-4 were to be used to mod games, it could be used to create new levels, characters, and items. It could also be used to change the rules of the game or add new features. This could potentially lead to a whole new level of creativity and innovation in the gaming world.

However, there are some challenges that would need to be overcome in order for GPT-4 to be used effectively to mod games. First, GPT-4 would need to be trained on a dataset of game code. This would allow it to understand the structure of games and how to modify them. Second, GPT-4 would need to be able to generate code that is compatible with the game engine that is being used. This would allow it to create mods that are actually playable.

Despite these challenges, it is possible that GPT-4 could be used to mod games in the future. If this were to happen, it could have a major impact on the gaming world. It could lead to a whole new level of creativity and innovation in the gaming world.

 

[BernL]

Just for fun I asked Google Bard.


It is possible that GPT-4 could be used to mod games in the future. GPT-4 is a large language model that has been trained on a massive dataset of text and code. This means that it has a deep understanding of language and how to use it to generate text, translate languages, write different kinds of creative content, and answer questions in an informative way.

If GPT-4 were to be used to mod games, it could be used to create new levels, characters, and items. It could also be used to change the rules of the game or add new features. This could potentially lead to a whole new level of creativity and innovation in the gaming world.

However, there are some challenges that would need to be overcome in order for GPT-4 to be used effectively to mod games. First, GPT-4 would need to be trained on a dataset of game code. This would allow it to understand the structure of games and how to modify them. Second, GPT-4 would need to be able to generate code that is compatible with the game engine that is being used. This would allow it to create mods that are actually playable.

Despite these challenges, it is possible that GPT-4 could be used to mod games in the future. If this were to happen, it could have a major impact on the gaming world. It could lead to a whole new level of creativity and innovation in the gaming world.

I suspect this will happen, could revolutionize gaming.

 

[BrunoBæ]

GPT-4 is a large language model that has been trained on a massive dataset of text and code. This means that it has a deep understanding of language

Understanding??
It doesnt understand a thing.
At least if understanding is defined as including something related to "awareness" of itself and "awareness" of some physical (or even internal) world.
The only thing this socalled IT is trained to is to "know" what letters does commonly stand beside of each other.
And when this is learned then what words does commonly stand beside of each other.

BæTheWay:"Commonly" in this context does only mean what letters and words STATISTICALLY does stand beside each other in related texts.

 

[demetri]

Understanding??
It doesnt understand a thing.
At least if understanding is defined as including something related to "awareness" of itself and "awareness" of some physical (or even internal) world.
The only thing this socalled IT is trained to is to "know" what letters does commonly stand beside of each other.
And when this is learned then what words does commonly stand beside of each other.

CatsAreTheWorstDogs: "Commonly" in this context does only mean what letters and words STATISTICALLY does stand beside each other in related texts.

Don't be so harsh, I think Andrew just copy-pasted a reply from Google AI here

 

[Deleted member 197115]

Arguing with and getting angry at AI means something is working.
But based on my limited questioning of this thing it's just stupid google gathered data gibberish generator. With no intelligence at all, just like typical amazon rep, the only thing it can probably successfully replace.

 

[BrunoBæ]

Don't be so harsh, I think Andrew just copy-pasted a reply from Google AI here

If it can be understood harsh then I appologize.
It wasnt ment so.
But the reason for the (misplaced) harsh-ish thing is probably because Im becomming slightly "tired" of all the people who believe we have some kind of "real" intelligence and understanding here.

BæTheWay: Until some weeks ago I was also tempted myself to believe the hype that something new and intelligent-ish was on the table. But then I heard a lecture of a scientist involved in exactly the development of socalled Artificial Intelligence. And after his correction and my following disappointment then Im slightly tired of hearing how humanlike and intelligent this socalled AI is.

 

[blekenbleu]

some kind of "real" intelligence

AI can do correlations, identifying patterns.
Attributing "real" intelligence to humans is IMO generous.

 

[BrunoBæ]

The only thing this socalled IT is trained to is to "know" what letters does commonly stand beside of each other.
And when this is learned then what words does commonly stand beside of each other.

The dynamic of exactly the rather unintelligent algoritms Im talking about will find what you call correlations and identifying patterns.
The only thing that is needed is statistics and a lot of data to be ordered by this statistics.

BæTheWay: Be aware that what you call finding correlations and patterns is just our human interpretation of this proces of statistically ordering of a lot of data.
Hehe and presenting it as an outcome/answer.

 

[BrunoBæ]

Be aware that the "Hehe" thing is not to be understood as patronizing.
If people want to believe that this GPT4 algoritm is "really" intelligent then no problem

 

[Deleted member 197115]

If it were really so dumb, Musk and other tech gurus wouldn't be so concerned.

 

[BrunoBæ]

Ahh come on. Do you REALLY think these bunch of overly rich guys are conserned about the civil society?
I guess even if a former president from US did know anything about IT then he would be part of this oh so conserned group too.

BæTheWay: The only reason they want to pause the development of AI is because they (so-oder-so) hope that at least some of their competitors will be slowed down.
Just check some of the critical IT journalists assessment of this pause joke.

 

[Deleted member 197115]

Yeah, and this is why Italy banned ChatGPT and Germany is working on following suite.
And these "rich guys" are 1,100 AI researchers and prominent tech leaders, including Tesla CEO and OpenAI co-founder Elon Musk and Apple co-founder Steve Wozniak.

 

[demetri]

If it were really so dumb, Musk and other tech gurus wouldn't be so concerned.

Musk is absolutely not a tech guru. He's a very talented (and lucky so far) BS artist. Maybe he's just afraid of the new competition to himself

 

[Deleted member 197115]

Musk is absolutely not a tech guru. He's a very talented (and lucky so far) BS artist. Maybe he's just afraid of the new competition to himself

He can be extravagant, flamboyant, and not always practical, but the guy contributed to tech progress in one way or another more than anyone in 21st century.
And speaking of AI, he was among pioneers and co-founder of OpenAI, ChatGPT creator.

 

[BrunoBæ]

Yeah, and this is why Italy banned ChatGPT

It really looks like you does look everywhere for something that can be used as an "argument" for saying that this rather unintelligent chatbot is so dangerously intelligent.
Because the reason Italy has blocked it is absolutely not because they are afraid of this dangerous artificial intelligences impact on civil society and human thinking.

Italy temporarily blocks ChatGPT over privacy concerns
Italy is temporarily blocking the artificial intelligence software ChatGPT in the wake of a data breach as it investigates a possible violation of stringent European Union data protection rules.

BæTheWay: When this chatbot does search the internet it supposedly are unable to obey some (probably commercial) standard protection rules

BæTheWay2: But lets stop this non-sence quarreling. Because as I said above believing in any deeper intelligence in this rather simple chatbot is ok with me.

 

[demetri]

He can be extravagant, flamboyant, and not always practical, but the guy contributed to tech progress in one way or another more than anyone in 21st century.
And speaking of AI, he was among pioneers and co-founder of OpenAI, ChatGPT creator.

Are you serious? He didn't actually found Paypal or Tesla (he won a lawsuit after he ousted the two guys who started the company, so now he can officially call himself a "founder"), or invented SpaceX reusable rockets. But he's been very good at making people believe that's all him all the way. You maybe can call him a visionary (though he falls short on basically all of his big promises and some of his ideas like underground highways are just plain stupid), but he's not a tech guru by any stretch of imagination. I gotta say he's been very good in exploiting various government-created business opportunities (green credits, electric vehicle tax rebates, government space contracts, etc) and his successful businesses probably would've failed without them. His contribution to AI was investing his money (among many others), not working on the actual technology. That is called an investor.

 

[Deleted member 197115]

Did your Tesla catch fire or something, why such a beef with the guy?

 

[BrunoBæ]

so now he can officially call himself a "founder"), or invented SpaceX reusable rockets.

I admit that for a long time I actually thought that one of Musks companies independently invented and created the SpaceX rockets.
But then I found out (by searching) that NASA was closely involved both in the tech part but also massively in the economically part.

BæTheWay: Probably as a way to sidestep/bypass some of the government restrictions of how much money NASA could have and use.