Benchmark design, contamination, harnesses, construct validity.https://securesein.com/https://securesein.com/blog/catching-ai-liars/https://securesein.com/blog/catching-ai-liars/Nineteen teams spent a month building AI lie detectors. The winning trick turned out to be less clever than it looked — and that's the interesting part.Sat, 12 Sep 2026 00:00:00 GMTSebastiaan with AIResearch — Deep diveai-safetyevaluationllmshttps://securesein.com/blog/llms-model-a-harsher-world/https://securesein.com/blog/llms-model-a-harsher-world/A new benchmark shows language models know what's wrong but not how people actually react to it — and the bias may come from alignment itself.Wed, 09 Sep 2026 00:00:00 GMTSebastiaan with AIResearch — Paperllmsai-safetyevaluation