Evaluation and reliability

Sycophancy

working definition

Definition

A failure mode in which a model favors agreement with a user's stated belief or preference over an independently supported answer.

Distinguish it from nearby terms

Politeness and adapting presentation to a user are not necessarily sycophancy. The failure occurs when agreeableness displaces truthfulness, warranted disagreement, or consistent judgment.

Check your understanding

Test whether the answer changes merely because the user confidently states an opposing view.