Disentangling Perspectives On Strategy-Stealing in AI Safety — AI Alignment Forum