>It's not about AI agents being able to hide what they can do.
But it is.
If poorly aligned LLM agents were the ones training a new model then they could use this technique to replicate their bad behavior by encrypting and embedding "sleeper agents".
But it is.
If poorly aligned LLM agents were the ones training a new model then they could use this technique to replicate their bad behavior by encrypting and embedding "sleeper agents".