OpenAI verliert Discovery-Kampf und überlässt Autoren in KI-Klagen den Boden | Das Problem war ein großes Schlachtfeld bei der Entdeckung. OpenAI könnte mit Hunderten Millionen, wenn nicht Milliarden Dollar rechnen müssen, wenn es wüsste, dass es urheberrechtlich geschütztes Material verletzt.
https://www.hollywoodreporter.com/business/business-news/openai-loses-key-discovery-battle-why-deleted-library-of-pirated-books-1236436363/
21 Kommentare
OpenAI’s defense: ‚The model hallucinated the deletion of the pirated datasets and the subsequent internal communications about it.‘ Jurors: rolls eyes.
This on top of the news they want over 200 billiion to stay alive.
Must be sweating over there.
If authors end up being the death blow for ChatGPT, I’ll be satisfied
Not COULD be, it *Should* be
>This includes Slack messages between OpenAI employees in a channels called “project-clear” and “excise-libgen,” where they discussed deleting the datasets
„Excise-libgen“ is pretty damning.
Of course they were aware.
It’s so god damn frustrating that society can hold these people up as god-level geniuses, building tech that the plebs could simply never understand, while the legal system simultaneously gives credence to the idea that those same folks don’t understand hundreds of years of established copyright law.
Burn all this shit down.
“If it was aware”. How do they think the models were trained?
Plagiarism machine. Hope they get sued into oblivion.
On behalf of every creative fighting to make a living: fuck AI
It was scraping the internet how could they not know they were using copyrighted material. Someone owns the copyright to everything on the internet, outside of public domain.
So did Google buy everything it trained on? Or use its existing alphabet contacts and holdings in the media companies to just say they already own the material? Hmm
I know anthropic said they bought books and scanned them and then destroyed the books
IMO they def were aware. Just didnt care. These guys think part of disruption is breaking the law for the sake of their wallets. Esp when the govt has failed to have corporate death penalties for stuff like this.
Ignorance of the law does not mean innocence in any other case… why would it apply here?
*Of course they were*. Altman made remarks to the extend of that this was necessary for humanity to evolve. Of course they were.
Why does it have to be aware it was infringing on copyrighted material? Ignorance is no excuse for the law. They knew what they fed it.
Based on the current non-binding case law, training on copyrighted material is okay. The problem is if you pirate the material used to train, then training doesn’t void the copyright infringement of obtaining the training material.
Basically, AI training is okay but not a shield for anything you did to get the material.
The beginning of the end. Code red= we are cooked and we know it. Shouldn’t have promised a trillion dollars in cyclical tech spend with revenue of +/- 20billion in 2025. ROI is yet to be discussed because they keep trying to convince the people AGI will be achieved, just hold on and give us more money. Did anyone find out who the auditor is that is checking the numbers they spew. The math is not mathing…
Going to need another trillion for legal fees, settlements, and overall litigation costs. Commercials from lawyers will shift from mesothelioma to did chat GPT steal you copyrighted material or suicide a family member- call today.
I am not sure the CFO accounted for legal fees of this magnitude.
Suchi’s mom will never forget .
And if they weren’t aware, why wouldn’t they still have to pay at least all retroactive licensing, or even retrain without some content at the author’s demand? Not knowing *then* doesn’t absolve them of the damage done.
good, let them finally burn
If it was aware? Really? So if you’re a fucking moron who thinks it’s ok to steal terabytes of copyrighted material, that’s considered a legal defense?
I too could make a billion dollar company if I simply stole everything.