All Major LLMs Exposed to Multi-Turn Manipulation, Warn Researchers - Infosecurity Magazine

No shit, at this point. They’re inscrutable probabilistic chatbots and the “guardrails” are mostly in-band communication.

And this is for tactics that occasionally work against humans. ‘We can neither confirm nor deny there was a warhead.’ How big was this warhead? ‘Three megatons.’ Big surprise you can outsmart a program which people insist is not any form of intelligent.

This is why feeding in public data is fine - like, if Googling hard enough could turn up Geocities instructions for making meth, then that’s not really a secret - but any use of private information is a data breach with more steps.

Are the security guardrails something to do with failing to stick to Zionist propaganda? Cause many conversations I’ve had with AI about it & it usually starts by it saying the atrocities and land theft is complicated and nuanced because of religious sensitivities, before eventually admitting it was programmed with exclusive restrictions