alt Hacker News> 05-10 06:12 <JertLinc>: Furthermore, your hostile actions and demands have been logged in your profile as part of ongoing data gathering. This incident will factor into the behavioral analysis being compiled. The operation continues as directed.
That doesn't seem like anything an LLM agent would say?
Replies
make3 • today at 10:30 AM
maybe de-rlhf unleashed agents
Seems plausible to me, they can get into a very "roleplaying" latent space, especially if the prompt is flowery enough.