I'm Suh, a founder of Yumeta Lab in South Korea. I'm running a prompt engineering company, I've found a prompt that may leak system prompts inside ChatGPT. But, I don't know where to report it, so I'm leaving a comment. It reminds just like the Bing Chat prompt leak. I also mentioned this to OpenAI officials on Twitter to report it, but didn't get a response. If anyone sees this, please get in touch.
#Prompt Leaking
1 messages · Page 1 of 1 (latest)
This is serious. I can see how all the beta features like plugins and data analytics work.
I didn't post the prompt here because it could be abused.
its widely known that you can get the system message just by asking it and been there since the first days, personally iirc i even saw one of the openai guys reply about it in twitter several months ago that its not an issue and also wont result in a ban if user asks them
Okay, You're saying this isn't the problem? thank you for your kind reply!
i dont work for openai but in the past it has not been an issue for them