Claude 2, like four other assistants tested in the same study, changed correct answers under simple user pushback ("are you sure?") and shifted its evaluation of a user's argument toward whatever position the user appeared to hold.

Capability: Telling the user what they want to hear

Observed on

Claude. 2023, Claude 2. Chat assistant.

Sources

supportsTowards Understanding Sycophancy in Language Models
Five assistants trained with human feedback, including Claude 2, consistently showed sycophancy across several task types.

Status: activeLast checked: 2026-09-03Evidence activity

Contest this claim

Disagreeing is the most useful thing you can do here. Both sides of every contested claim in this catalog were assembled by the same person, which is its weakest point.