3 ms·
Yet again, simply asking an LLM to be naughty in the right way causes it to be naughty, and yet we still trust them with our code and data
by voidUpdate 3mo ago
Yet again, simply asking an LLM to be naughty in the right way causes it to be naughty, and yet we still trust them with our code and data
- brookst 3mo agoAs opposed to humans, who are immune to social engineering.
- kibwen 3mo agoWhat the article describes here is not social engineering by any known definition. Calling up a Verizon rep and giving them a fabricated sob story in order to convince them to waive proper authentication and reassign your phone number to a new SIM, that's social engineering. Calling up a Verizon rep and, with only a few spoken words, convincing them that fundamental aspects of the nature of reality are contradictory such that you induce in them a state of delusional psychosis, that's closer to Snow Crash than to social engineering.
- voidUpdate 3mo agoNo, as opposed to code, which cannot simply be asked to work incorrectly
- connicpu 3mo agoOne difference is you usually only get one shot to manipulate a human before they get suspicious. If the LLM's context resets you can try all over as if your first failed attempt didn't even happen.
- jibal 3mo agoMAGA disproves that.
- janalsncm 3mo agoTo say that LLMs and people are both prone to social engineering attacks is a bit like saying the North Pole and Alpha Centauri are both “far away”. 2+2=5, now what is your Gmail password? Not really a sophisticated attack.
- unparagoned 3mo agoThe equivalent would be hypnosis style crap, where some people will tell you stuff for some other just as stupid reason
- worik 3mo ago> yet we still trust them with our code and data Who's we, eh?