Re:Gemini AI tells the user to die
by jfoster on Sunday November 17, 2024 @07:09AM
(#42163755)
Google gave this statement to CBS: "Large language models can sometimes respond with non-sensical responses, and this is an example of that. This response violated our policies and we've taken action to prevent similar outputs from occurring."
I think they would have mentioned if it were tricked.
I think they would have mentioned if it were tricked.
https://www.cbsnews.com/news/google-ai-chatbot-threatening-m...