LLMs Can't Distinguish Between What They Know and Where They're Guessing

But then how could they possibly know that?

Google just figured out why AI lies with confidence.
Large language models still make confident mistakes on simple factual questions.
A new paper from Google Research explains why this keeps happening.
Models cannot reliably tell what they know from what they are guessing.… pic.twitter.com/CXPpQwqzfr

— AlphaSignal AI (@AlphaSignalAI) May 23, 2026