Sahwa@reddthat.com to Technology@lemmy.worldEnglish · 20 hours agoGoogle's Agentic AI wipes user's entire HDD without permission in catastrophic failurewww.tomshardware.comexternal-linkmessage-square121fedilinkarrow-up1578arrow-down112
arrow-up1566arrow-down1external-linkGoogle's Agentic AI wipes user's entire HDD without permission in catastrophic failurewww.tomshardware.comSahwa@reddthat.com to Technology@lemmy.worldEnglish · 20 hours agomessage-square121fedilink
minus-squareDevial@discuss.onlinelinkfedilinkEnglisharrow-up111arrow-down3·14 hours agoIf you gave your AI permission to run console commands without check or verification, then you did in fact give it permission to delete everything.
minus-squareVictor@lemmy.worldlinkfedilinkEnglisharrow-up3arrow-down1·4 hours agoBut for real, why would the agent be given the ability to run system commands in the first place? That sounds like a gargantuan security risk.
minus-squareutopiah@lemmy.worldlinkfedilinkEnglisharrow-up1·2 hours agoBecause “agentic”. IMHO running commands is actually cool, doing it without very limited scope though (as he did say in the video) is definitely idiotic.
minus-squarelando55@lemmy.ziplinkfedilinkEnglisharrow-up24arrow-down1·11 hours agoI didn’t install leopards ate my face Ai just for it to go and do something like this
If you gave your AI permission to run console commands without check or verification, then you did in fact give it permission to delete everything.
But for real, why would the agent be given the ability to run system commands in the first place? That sounds like a gargantuan security risk.
Because “agentic”. IMHO running commands is actually cool, doing it without very limited scope though (as he did say in the video) is definitely idiotic.
I didn’t install leopards ate my face Ai just for it to go and do something like this