• 1 Post
  • 25 Comments
Joined 2 years ago
cake
Cake day: July 28th, 2023

help-circle
rss


  • While you are correct that there likely is no intention and certainly no self-awareness behind the scheming, the researchers even explicitly list the option that the AI is roleplaying as an evil AI, simply based on its training data, when discussing the limitations of their research, it still seems a bit concerning. The research shows that given a misalignment between the initial prompt and subsequent data modern LLMs can and will ‘scheme’ to ensure their given long-term goal. It is no sapient thing, but a dumb machine with the capability to decive its users, and externalise this as shown in its chain of thought, when there are goal misalignments seems dangerous enough. Not at the current state of the art but potentially in a decade or two.





  • And we have better night vision than most the animals that have better day-vision than us. Humans are like the Leatherman of animals. Universally capable of doing most things but not as good as something specialized for that task. Plus of course capable of coming up with ways to cheat





  • Looks like some kind of intrusion of magma (the pale rock) into the darker rock. you can see how all the veins seem to originate from the pale rock, also the broken-of dark part in the pale rock seems to indicate it, could be part of the original rock, that surrounded the magma, before breaking of.

    Some actual geologist might want to give their opinion though. I only had like two years of geology at university before shifting my studies toward crystallography and crystal growth.




  • Depending on the way the fey phrased the request you may either have insulted it, or broken a deal. If it asks to have your name and you answer with “yes, my name is <fake name>.” You have lied to it and broken a simple transaction. It may even be justified in using ever more harsh measures to extract your name from you, as it has a legitimate claim on it. At least that is the way I would play such a situation for the fey. For me the thing is that trying to outsmart them rarely works, because they know netter in which rules they operate than you do. A answer like "You may not have my name, but you can call me <fake name>. " Should work better in my opinion. Sorry for the wall of text btw. Just being bored out of my mind rn.