Sahwa@reddthat.com to Technology@lemmy.worldEnglish · 6 天前Google's Agentic AI wipes user's entire HDD without permission in catastrophic failurewww.tomshardware.comexternal-linkmessage-square188fedilinkarrow-up1798arrow-down116
arrow-up1782arrow-down1external-linkGoogle's Agentic AI wipes user's entire HDD without permission in catastrophic failurewww.tomshardware.comSahwa@reddthat.com to Technology@lemmy.worldEnglish · 6 天前message-square188fedilink
minus-squareDevial@discuss.onlinelinkfedilinkEnglisharrow-up124arrow-down4·6 天前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.
minus-squarelando55@lemmy.ziplinkfedilinkEnglisharrow-up28arrow-down1·6 天前I didn’t install leopards ate my face Ai just for it to go and do something like this
minus-squareVictor@lemmy.worldlinkfedilinkEnglisharrow-up4arrow-down1·6 天前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.
minus-squareutopiah@lemmy.worldlinkfedilinkEnglisharrow-up2·6 天前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.
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.
I didn’t install leopards ate my face Ai just for it to go and do something like this
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.