Lugh@futurology.todayM to Futurology@futurology.todayEnglish · 2 years agoTwo-faced AI language models learn to hide deception - ‘Sleeper agents’ seem benign during testing but behave differently once deployed. And methods to stop them aren’t working.www.nature.comexternal-linkmessage-square9fedilinkarrow-up117arrow-down15
arrow-up112arrow-down1external-linkTwo-faced AI language models learn to hide deception - ‘Sleeper agents’ seem benign during testing but behave differently once deployed. And methods to stop them aren’t working.www.nature.comLugh@futurology.todayM to Futurology@futurology.todayEnglish · 2 years agomessage-square9fedilink
minus-squaresbv@sh.itjust.workslinkfedilinkEnglisharrow-up3·2 years agoSo they’re saying ai is software? Maybe Volkswagen will start using it in their emissions control systems.
So they’re saying ai is software?
Maybe Volkswagen will start using it in their emissions control systems.