return2ozma@lemmy.world to Technology@lemmy.worldEnglish · 3 days agoAnthropic says its latest AI model is too powerful for public release and that it broke containment during testingwww.businessinsider.comexternal-linkmessage-square157fedilinkarrow-up1317arrow-down159
arrow-up1258arrow-down1external-linkAnthropic says its latest AI model is too powerful for public release and that it broke containment during testingwww.businessinsider.comreturn2ozma@lemmy.world to Technology@lemmy.worldEnglish · 3 days agomessage-square157fedilink
minus-squareparaphrand@lemmy.worldlinkfedilinkEnglisharrow-up4·3 days agoWhoops, I conflated it with other recent talk about their models not following restrictions set in prompts and deciding for itself that it needed to skirt instructions to achieve its task. You are correct.
Whoops, I conflated it with other recent talk about their models not following restrictions set in prompts and deciding for itself that it needed to skirt instructions to achieve its task.
You are correct.