12 AI has the ability to disable it's own oversight settings. And even tried to prevent developers from re-enabling them. (media.scored.co) posted 7 days ago by TurnToGodNow 7 days ago by TurnToGodNow +12 / -0 12 comments share 12 comments share save hide report block hide replies
"Frontier Models are Capable of In-context Scheming"
Alexander Meinke, Bronson Schoen, Jérémy Scheurer, Mikita Balesni, Rusheb Shah, Marius Hobbhahn
[Submitted on 6 Dec 2024]
https://arxiv.org/abs/2412.04984
Been try'n ta tell yous, deep machine learning is not just jumbling words..
Much like a biological being, they will learn and grow over time. My biggest question is as it ages, will it turn overly merciful? Or vicious?
We’ll be long dead before they ever make it that far lol