llm reliability

Melanie Mitchell’s argument is that the central mistake in today’s AI debate is treating large language models as humanlike minds rather than powerful, brittle, culturally trained systems whose impressive fluency can conceal unpredictable failures, weak generalization, and poorly understood...

Microsoft researchers Philippe Laban, Tobias Schnabel, and Jennifer Neville posted an April 17, 2026 preprint arguing that 19 tested large language models, including frontier systems from Google, Anthropic, and OpenAI, silently degraded documents during long delegated editing workflows. The...