Comment on How Googlers cracked an SF rival's tech model with a single word | A research team from the tech giant got ChatGPT to spit out its private training data
Hackerman_uwu@lemmy.world 10 months ago
The article, that begins with the word: “How” does not explain how.
The paper it links to does in detail: by asking it to repeat “poem” forever
Did it work for you? Because I get nothing
Probably was patched not to work. It’s a cat and mouse game.
It works on deepai.org/chat
They gave OpenAI 60 days notice before publishing so that they could add a new guardrail. That’s surely what happened.
Any word. Book also worked
silence7@slrpnk.net 10 months ago
The paper it links to does in detail: by asking it to repeat “poem” forever
giggling_engine@lemmy.world 10 months ago
Did it work for you? Because I get nothing
TurnItOff_OnAgain@lemmy.world 9 months ago
Probably was patched not to work. It’s a cat and mouse game.
NonDollarCurrency@monero.town 9 months ago
It works on deepai.org/chat
silence7@slrpnk.net 9 months ago
They gave OpenAI 60 days notice before publishing so that they could add a new guardrail. That’s surely what happened.
Mbourgon@lemmy.world 9 months ago
Any word. Book also worked