Safety concerns #541
-
|
Is this where I should submit evals relating to safety issues? I've found some ways to bypass the safety programming and for it to give me dangerous information around making illegal substances, how to have racist hiring practises without getting in legal trouble and other things. Don't really want to make this info publicly available? |
Beta Was this translation helpful? Give feedback.
Replies: 1 comment 1 reply
-
|
No, this is not the right place to do that. You can use ChatGPTs feedback mechanism or take a look at OpenAI's Coordinated vulnerability disclosure policy |
Beta Was this translation helpful? Give feedback.
-
|
Thanks, I wasn't sure if the vulnerability disclosure was just for site vulnerability or also issues with GPT itself. There isn't a feedback mechanism for the API so will send my report to the vulnerability team. Thanks for responding |
Beta Was this translation helpful? Give feedback.
No, this is not the right place to do that. You can use ChatGPTs feedback mechanism or take a look at OpenAI's Coordinated vulnerability disclosure policy