Automating subtle emergent misalignment research

T2Covered by 1 source · first reported by AI Alignment Forum / LessWrong (curated) at 03 Nov, 21:01 UTC

Authors: Oscar Balcells Obeso, Richard Qi, Monte MacDiarmid, Evan Hubinger Abstract Red-teaming by humans is a vital part of model safety testing, but it may struggle to keep up as models improve. To test whether automated safety techniques could scale (which would allow us to turn more compute…