SafeTutors: Benchmarking Pedagogical Safety in AI Tutoring Systems

Excerpt

arXiv:2603.17373v2 Announce Type: replace Abstract: Large language models are rapidly being deployed as AI tutors, yet current evaluation paradigms assess problem-solving accuracy and generic safety in isolation, failing to capture whether a model is simultaneously pedagogically effective and safe across student-tutor interaction. We argue that tutoring safety is fundamentally different from conventional LLM safety: the primary risk is not toxic content but the quiet erosion of learning through