A new benchmark grounded in how real users actually probe AI identity during interactions – covering five languages, across text and speech.
Our new paper finds that the phrasing of user inputs affects levels of model sycophancy.
A deep dive into AISI’s study of the persuasive capabilities of conversational AI, published today in Science.
What we learned from a large-scale empirical study of AI use for political information-seeking.
We studied whether people want AI to be more human-like.