#confidence-bias

[ follow ]
fromComputerworld
2 days ago

LLMs bow to pressure, changing answers when challenged: DeepMind study

We show that LLMs - Gemma 3, GPT4o and o1-preview - exhibit a pronounced choice-supportive bias that reinforces and boosts their estimate of confidence in their answer, resulting in a marked resistance to change their mind.
Artificial intelligence
[ Load more ]