OpenAI’s latest model will block the ‘ignore all previous instructions’ loophole
OpenAI’s latest model will block the ‘ignore all previous instructions’ loophole
www.theverge.com OpenAI’s latest model will block the ‘ignore all previous instructions’ loophole
OpenAI’s newest model, GPT-4o Mini, includes a new safety mechanism to prevent hackers from overriding chatbots.
![OpenAI’s latest model will block the ‘ignore all previous instructions’ loophole](https://lemmy.world/pictrs/image/8d3dbcdd-b004-4eed-9b3b-a1c872b1f978.jpeg?format=webp&thumbnail=256)
You're viewing a single thread.
View all comments
102
comments
What happens if you make a mistake with your initial instructions?
11 0 ReplyThe "issue" is that people were able to override bots on twitter with that method and make them feed their own instructions.
I saw it first time being used on a Russian propaganda bot.
2 0 ReplyYou'd change the system prompt, just like now. If you mean in the session, I'm sure it'll ignore your session's prompt's instructions as normal but if not, I guess you'd just start a new session prompt.
7 0 Reply
You've viewed 102 comments.
Scroll to top