After GPT-4o backlash, researchers benchmark models on moral endorsement—Find sycophancy persists across the board https://t.co/Wh8Zel3R9q on https://twitter.com/AcerboLivio/status/1925816548842676268
a free community where sharing your ideas!