Re:Show HN: The largest collection of ChatGPT jailbreaks on ...
by nextaccountic on Wednesday March 01, 2023 @10:21AM
(#34983320)
The AI could be built differently, it's just that the current LLM trend makes AI train to guess the continuation of the text, which means it says what it guesses you expect it to say
[ Parent
| on HN ]
There are no comments on this comment yet.