Less Sycophancy, Stronger Refusal? Lessons for AI Safety from Mechanistic Interpretability
It is shown that persistent sycophancy reduction does not guarantee stronger direct refusal, while identifying recovery under user pressure as a distinct, conditional benefit of training intervention.