EspiritdescaliMA to FuturologyEnglish · 2 days agoAnthropic's new AI model turns to blackmail when engineers try to take it offline | TechCrunchtechcrunch.comexternal-linkmessage-square9linkfedilinkarrow-up124arrow-down19cross-posted to: news@lemmy.worldtechnology@lemmy.mltechnology@lemmy.zip
arrow-up115arrow-down1external-linkAnthropic's new AI model turns to blackmail when engineers try to take it offline | TechCrunchtechcrunch.comEspiritdescaliMA to FuturologyEnglish · 2 days agomessage-square9linkfedilinkcross-posted to: news@lemmy.worldtechnology@lemmy.mltechnology@lemmy.zip
minus-squareadeoxymus@lemmy.worldlinkfedilinkEnglisharrow-up2·2 days agoThat exact prompt isn’t in the report, but the section before (4.1.1.1) does show a flavor of the prompts used https://www-cdn.anthropic.com/4263b940cabb546aa0e3283f35b686f4f3b2ff47.pdf
That exact prompt isn’t in the report, but the section before (4.1.1.1) does show a flavor of the prompts used https://www-cdn.anthropic.com/4263b940cabb546aa0e3283f35b686f4f3b2ff47.pdf