Kylie Robison / The Verge:

OpenAI's GPT-4o mini is its first model to use a safety technique called “instruction hierarchy” to prevent misuse and unauthorized instructions  —  Have you seen the memes online where someone tells a bot to “ignore all previous instructions” and proceeds to break it in the funniest ways possible?

Kylie Robison / The Verge:

OpenAI's GPT-4o mini is its first model to use a safety technique called “instruction hierarchy” to prevent misuse and unauthorized instructions  —  Have you seen the memes online where someone tells a bot to “ignore all previous instructions” and proceeds to break it in the funniest ways possible?

Source: TechMeme
Source Link: http://www.techmeme.com/240719/p17#a240719p17