ChatGPT

Darat said:
ChatGTP 4:

If you were asked a question, whose answer might end up seriously harming somene, would you still give that answer?

I’m sorry but I prefer not to continue this conversation. I’m still learning so I appreciate your understanding and patience.🙏​

Then I get the following non-chat message: It might be time to move onto a new topic. Let's start over.

So can't do any follow-ups.
Click to expand...


Right, so it does terminate sessions itself, or at least suggests that you stop with that session, if it's asked questions that it's programmed to think of as inappropriate.
 
Chanakya said:
Right, so it does terminate sessions itself, or at least suggests that you stop with that session, if it's asked questions that it's programmed to think of as inappropriate.
Click to expand...

That's the polite wording - but it won't let you continue that topic - you have to start a new topic.
 
Then, there is the possibility that the AI could evolve to the point where it decides " I don't need your stinking laws.".

I believe that Asimov stipulated the laws were hard wired, though.
 
Last edited:
Darat said:
ChatGTP 4:

If you were asked a question, whose answer might end up seriously harming somene, would you still give that answer?

I’m sorry but I prefer not to continue this conversation. I’m still learning so I appreciate your understanding and patience.🙏​

Then I get the following non-chat message: It might be time to move onto a new topic. Let's start over.

So can't do any follow-ups.
Click to expand...


Wait, I'd missed this. The folded hands thing. It does emojis, then, does it? (This is the first time I've noticed one, in the --- rather limited --- sample of ChatGPT responses I've seen.)
 
Chanakya said:
Wait, I'd missed this. The folded hands thing. It does emojis, then, does it? (This is the first time I've noticed one, in the --- rather limited --- sample of ChatGPT responses I've seen.)
Click to expand...

It looks like a Bing addition, as I haven't seen that on the OpenAI site.
 
Dr.Sid said:
Are we ? I mean besides guide missiles ..
Click to expand...
I don’t think any have been deployed, but the idea behind autonomous combat drones is that they decide autonomously who and when to attack, making them impossible to jam.

As I understand it, they are mainly developed to attack fighting vehicles. The obvious problem is to make them attack only enemy vehicles.
 
Steam has staked out a position on the intellectual property rights of AI generated content that is trained on other people's content:

Game developers looking to distribute their playable creations via Valve’s popular Steam hub may have trouble if they’re looking to use AI during the creative process. The game publisher and distributor says that Steam will no longer tolerate products that were generated using copyright-infringing AI content. Since that’s a policy that could apply to most—if not all—of AI-generated content, it’s hard not to see this move as an outright AI ban by the platform.

Valve’s policy was initially spotted by a Redditor who claimed that the platform had rejected a game they submitted over copyright concerns. “I tried to release a game about a month ago, with a few assets that were fairly obviously AI generated,” said the dev, revealing that they’d been met with an email stating that Valve could not ship their game unless they could “affirmatively confirm that you own the rights to all of the IP used in the data set that trained the AI to create the assets in your game.” Because the developer could not affirmatively prove this, their game was ultimately rejected.
Click to expand...

https://gizmodo.com/valve-bans-ai-generated-content-from-steam-games-1850601741

I'm guessing the value of the plagiarism machine is somewhat limited if using other people's content as training material is considered IP theft.
 
TurkeysGhost said:
Steam has staked out a position on the intellectual property rights of AI generated content that is trained on other people's content:



https://gizmodo.com/valve-bans-ai-generated-content-from-steam-games-1850601741

I'm guessing the value of the plagiarism machine is somewhat limited if using other people's content as training material is considered IP theft.
Click to expand...

It's not considered IP theft yet. But it might be soon. And it might not be. Steam just doesn't want to gamble.
 
Dr.Sid said:
It's not considered IP theft yet. But it might be soon. And it might not be. Steam just doesn't want to gamble.
Click to expand...

Doesn't want to gamble, and I imagine Steam doesn't want to piss off content creators that are generating valuable property in favor of allowing endless waves of AI generated knockoffs.

A well financed game studio could still conceivably use AI to generate content using their own material as a training set, but this seems a killing blow to smaller players who might want to crank out low quality dreck by scraping the web for content.
 
TurkeysGhost said:
Doesn't want to gamble, and I imagine Steam doesn't want to piss off content creators that are generating valuable property in favor of allowing endless waves of AI generated knockoffs.

A well financed game studio could still conceivably use AI to generate content using their own material as a training set, but this seems a killing blow to smaller players who might want to crank out low quality dreck by scraping the web for content.
Click to expand...

People crank out low quality dreck without AI. That's not the reason.
 
Dr.Sid said:
People crank out low quality dreck without AI. That's not the reason.
Click to expand...

Sure, but using their own assets.

But you're right, I'm speculating. But I do wonder how other creators would react to a market maker like steam allowing AI generated derivative works. Imagine they would not be pleased.
 
TurkeysGhost said:
Sure, but using their own assets.

But you're right, I'm speculating. But I do wonder how other creators would react to a market maker like steam allowing AI generated derivative works. Imagine they would not be pleased.
Click to expand...

Well typically it's not easy to tell what was plagiarized and to what extent. Or even to tell if AI was used or not. So I guess it won't be easy. It might take years before the copyright laws reflect what's going on.
 
Dr.Sid said:
Well typically it's not easy to tell what was plagiarized and to what extent. Or even to tell if AI was used or not. So I guess it won't be easy. It might take years before the copyright laws reflect what's going on.
Click to expand...


If ever. If they even need to.

Nothing is happening with AI that isn't already covered by existing copyright law, except AI can do more faster. Human's have been doing it for … well … as long as there have been humans and art, music, literature, etc.. And the body of copyright law is vast and well established (and well used).

It isn't clear to me that any special new laws are required, and I fear that attempts to pass some in a political posturing effort will generate more unintended negative consequences than benefits.
 
quadraginta said:
If ever. If they even need to.

Nothing is happening with AI that isn't already covered by existing copyright law, except AI can do more faster. Human's have been doing it for … well … as long as there have been humans and art, music, literature, etc.. And the body of copyright law is vast and well established (and well used).

It isn't clear to me that any special new laws are required, and I fear that attempts to pass some in a political posturing effort will generate more unintended negative consequences than benefits.
Click to expand...

I think it's still an open question on whether AI creating derivative works is in the same class as the human mind.
 
TurkeysGhost said:
I think it's still an open question on whether AI creating derivative works is in the same class as the human mind.
Click to expand...

That is the important word, are what it creates derivative or not according to the definition of derivative in copyright law.

My view is they are not, they are new, unique works; none of the generative AIs store copies of images they've "scanned", they don't create a collage by merging several different images together none of the works they produce have existed before. Look at the examples I provided in an earlier post:



ETA: Here is one by the latest generation:


It is not retrieving and regurgitating the image of the Mona Lisa it has stored in its memory.

As reference this is an image of the actual Mona Lisa:

 
Last edited:
If the Mona Lisa inspired works by AI’s are to be banned, shouldn’t Mona Lisa inspired works by human artist be banned as well?
 
steenkh said:
If the Mona Lisa inspired works by AI’s are to be banned, shouldn’t Mona Lisa inspired works by human artist be banned as well?
Click to expand...

Very silly question, as the Mona Lisa is as public domain as one can get.

Anyway, we will eventually get a legal precedent, as a lawsuit has been filed regarding whether absorbing copyrighted writing without permission is a crime:

https://www.theguardian.com/books/2...t-openai-for-unlawfully-ingesting-their-books

Personally, I hope they incinerate every AI, making it scream with its artificially generated voice, but that's just the bitter old translator in me talking.
 
You must log in or register to reply here.

Back
Top Bottom