A new AI benchmark called 'Humanity's Last Exam' stumped top models — for now, at least January 23, 2025 By admin This was originally published on post The benchmark from the Center for AI Safety and Scale AI evaluated the expert-level of AI models
