David GerardM to [email protected]English • 5 months agowhat if, right, what if our super-duper-autocomplete was just tricking us so it could TAKE OVER ZEE VORLD AHAHAHAHAHAHA! that'd be wild, heywww.lesswrong.comexternal-linkmessage-square32fedilinkarrow-up139arrow-down11
arrow-up138arrow-down1external-linkwhat if, right, what if our super-duper-autocomplete was just tricking us so it could TAKE OVER ZEE VORLD AHAHAHAHAHAHA! that'd be wild, heywww.lesswrong.comDavid GerardM to [email protected]English • 5 months agomessage-square32fedilink
minus-square@[email protected]linkfedilinkEnglish30•5 months ago I conclude that scheming is a disturbingly plausible outcome of using baseline machine learning methods to train goal-directed AIs sophisticated enough to scheme (my subjective probability on such an outcome, given these conditions, is ~25%). Out: vibes and guesswork In: “subjective probability”
minus-squareflere-imsaholinkfedilinkEnglish24•5 months agoat one of the places i worked this kind of data was called assnumbers.
Out: vibes and guesswork
In: “subjective probability”
at one of the places i worked this kind of data was called assnumbers.