The ‘safeguards’ of AI are funny, as they are manually put in by humans on certain phrases and token triggers. By rephrasing it you trigger a whole new token chain that bypasses the known ones.
Or, you know, get any of the hundreds of models you can self host that don’t have any of the safeguards in place or can be broken.
In the end it matters very little. It actually kinda sounds like the the same kind of fear mongering when Google became useful in the 2000s.
Now everyone can find the Anarchist Cookbook! Why won’t anybody think of the children! Terrorism! Rabble, rabble rabble.
My favorite jail break was using ASCII art text letters to get around the system prompt. Whoever came up with that probably died due to not being able to stop laughing.
The ‘safeguards’ of AI are funny, as they are manually put in by humans on certain phrases and token triggers. By rephrasing it you trigger a whole new token chain that bypasses the known ones.
Or, you know, get any of the hundreds of models you can self host that don’t have any of the safeguards in place or can be broken.
In the end it matters very little. It actually kinda sounds like the the same kind of fear mongering when Google became useful in the 2000s.
Now everyone can find the Anarchist Cookbook! Why won’t anybody think of the children! Terrorism! Rabble, rabble rabble.
My favorite jail break was using ASCII art text letters to get around the system prompt. Whoever came up with that probably died due to not being able to stop laughing.