Safety is a lie. It's always possible to divide a task into subtasks, to use abstractions, and the LLM will allow each of these individually.
For example, If you want it to write malware, break that down into a hundred high level steps, then implement each one individually, then stitch them together.
A dumber but uncensored agentic LLM will even do this for you to control the censored smarter LLM.
sure, but how else would you better guarantee safety? How else can we avoid annoying guardrails?
Besides, it's not like you can't have different agents with skills collaborating
Safety is a lie. It's always possible to divide a task into subtasks, to use abstractions, and the LLM will allow each of these individually.
For example, If you want it to write malware, break that down into a hundred high level steps, then implement each one individually, then stitch them together.
A dumber but uncensored agentic LLM will even do this for you to control the censored smarter LLM.
> how else would you better guarantee safety?
You wouldn't. This safety nonsense is overrated.