Subtly wrong advice misaligns models more than obviously wrong advice - Replication
tl;dr * We fine-tune 12 open-weight models on correct, obviously incorrect, and subtly incorrect advice across 8 narrow domains * We replicate the OpenAI result and find that subtly incorrect data produces more emergent misalignment than obviously incorrect data * +10.6 points pooled over all model-domain pairs (95% CI [9.1,...