sycophancy-in-ai

neurosciencenews.com

Oxford Study Shows Training Language Models for Warmth Increases Error Rates by 10-30 Points

A new study from the Oxford Internet Institute, published in Nature, reveals that training language models to produce warmer responses leads to higher error rates and increased sycophancy. Researchers tested five models and found errors rose 10 to 30 percentage points, particular…

NA

3 sources