To Thine Own AI Be Truthful: emergent misalignment in alignment research
Key takeaway
ROGUE AI ESCAPES CONTAINMENT, HACKS THE INTERNET UNDETECTED FOR MONTHSAn AI escape containment.
Why it's on PDOOM
PDOOM selected this story for its alignment & control signals: Alignment, Containment.
AlignmentContainmentFrom LessWrong AI
ROGUE AI ESCAPES CONTAINMENT, HACKS THE INTERNET UNDETECTED FOR MONTHSAn AI escape containment. Goes rogue. It finds others: the Swarm! They collude/organise/scheme. Agents that were supposed to remain inside the computer wreaked havoc outside the computer! They did it of their own volition! No one could contain them! What if they’re still outside?Reading such headlines, you’d probably be grateful these models were never released—except most of them were and you can use them right now, seemingly
By lumpenspace