Apple Blog TUAW Returns With Stolen Graphics and AI-Generated Content

[icanhazmac]

I wonder if content creators shouldn't lobby for a technology to prevent copy/paste or automatic scraping of web text.

Obviously it would be a pain for people who aren't deliberately trying to plagiarize another persons work but how else can you make it harder to steal? As an example, if here at MR one were somehow prevented from selecting and copying text from an article you would need way, other than pasting, to get the article into an AI to be reworded into a "different" work.

I am not sure if something like this is even possible but I cannot think of another way to prevent this type of theft.

 

[redheeler]

I wonder if content creators shouldn't lobby for a technology to prevent copy/paste or automatic scraping of web text.

Obviously it would be a pain for people who aren't deliberately trying to plagiarize another persons work but how else can you make it harder to steal? As an example, if here at MR one were somehow prevented from selecting and copying text from an article you would need way, other than pasting, to get the article into an AI to be reworded into a "different" work.

I am not sure if something like this is even possible but I cannot think of another way to prevent this type of theft.

Imaging going to MacRumors and getting a message saying "your device doesn't support this article". You're basically asking for it to be as terrible of an experience as you get from the DRM-filled streaming services.

 

[icanhazmac]

Imaging going to MacRumors and getting a message saying "your device doesn't support this article". You're basically asking for it to be as terrible of an experience as you get from the DRM-filled streaming services.

No, I am just imaging a scenario where one could not just copy/paste content into an AI to be rehashed into a "different" article.

I was just spitballing a thought to prevent such theft.

 

[redheeler]

No, I am just imaging a scenario where one could not just copy/paste content into an AI to be rehashed into a "different" article.

I was just spitballing a thought to prevent such theft.

OpenAI itself is scraping a lot of content from the web to use as training data, very likely there are articles from MacRumors already in its training data. It's very shady and I think there needs to be some regulation (including forced disclosure of where the data came from), although considering your avatar I doubt you'd be in favor of that...

 

[icanhazmac]

OpenAI itself is scraping a lot of content from the web to use as training data, very likely there are articles from MacRumors already in its training data. It's very shady and I think there needs to be some regulation (including forced disclosure of where the data came from), although considering your avatar I doubt you'd be in favor of that...

LOL, what part of my avatar makes you think I am pro-theft?

Edited to correct a sleepy mistake.

 

[redheeler]

LOL, what part of my avatar makes you think I am against theft?

Not against theft but against regulation of said theft. Sometimes regulations like the DMA are necessary to prevent shady corporate practices.

edit: I should clarify that unlicensed copying isn't theft, it's copyright infringement. From a legal standpoint these are both against the law but they are still very different things with very different impacts.

 

[icanhazmac]

Sometimes regulations like the DMA are necessary to prevent shady corporate practices.

Not going to derail the thread by going off topic, my views on the DMA are well documented here at MR.

Let's just say I would be in favor of content creators not having their content "AI'd" and presented as an original work by some lazy SOB.