Between a Rock and a Hard Place: The Tension Between Ethical Reasoning and Safety Alignment in LLMs

ArXi:2509.05367v4 Announce Type: replace-cross Large Language Model safety alignment predominantly operates on a binary assumption that requests are either safe or unsafe. This classification proves insufficient when models encounter ethical dilemmas, where the capacity to reason through moral trade-offs creates a distinct attack surface. We formalize this vulnerability through TRIAL, a multi-turn red-teaming methodology that embeds harmful requests within ethical framings.