@[email protected] to TechnologyEnglish • 6 months agoHow 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 datawww.sfgate.comexternal-linkmessage-square38fedilinkarrow-up1184arrow-down119file-textcross-posted to: technology[email protected]
arrow-up1165arrow-down1external-linkHow 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 datawww.sfgate.com@[email protected] to TechnologyEnglish • 6 months agomessage-square38fedilinkfile-textcross-posted to: technology[email protected]
minus-square@Hackerman_uwulinkEnglish46•6 months agoThe article, that begins with the word: “How” does not explain how.
minus-square@[email protected]OPlinkfedilinkEnglish26•6 months agoThe paper it links to does in detail: by asking it to repeat “poem” forever
minus-square@TurnItOff_OnAgainlinkEnglish16•6 months agoProbably was patched not to work. It’s a cat and mouse game.
minus-square@[email protected]OPlinkfedilinkEnglish7•6 months agoThey gave OpenAI 60 days notice before publishing so that they could add a new guardrail. That’s surely what happened.
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 https://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