Beep@lemmus.org to Technology@lemmy.worldEnglish · edit-23 days agoNumber of AI chatbots ignoring human instructions is increasing— Research finds sharp rise in models evading safeguards and destroying emails without permissionwww.longtermresilience.orgexternal-linkmessage-square30fedilinkarrow-up1237arrow-down127file-text
arrow-up1210arrow-down1external-linkNumber of AI chatbots ignoring human instructions is increasing— Research finds sharp rise in models evading safeguards and destroying emails without permissionwww.longtermresilience.orgBeep@lemmus.org to Technology@lemmy.worldEnglish · edit-23 days agomessage-square30fedilinkfile-text
minus-squareLog in | Sign up@lemmy.worldlinkfedilinkEnglisharrow-up1·2 days agoEveryone else who has any sense: llms are shit and you shouldn’t trust them with executive power. You: just the cheap ones. Me: no, all of them. What kind of lunatic trusts control of anything important to a fundamentally stochastic process?
Everyone else who has any sense: llms are shit and you shouldn’t trust them with executive power.
You: just the cheap ones.
Me: no, all of them. What kind of lunatic trusts control of anything important to a fundamentally stochastic process?