Consider the fine tuning already being done to build safety guardrails.. this is the evil version of...

infinite_spin • today at 4:22 AM • 1 reply • view on HN

Consider the fine tuning already being done to build safety guardrails.. this is the evil version of that.

shepherdjerred • today at 4:58 AM

If you can align an agent towards "good" then it's reasonable to think you can align it towards "evil"

alt Hacker News