AI glossary · Ethics, safety and privacy
AI sycophancy
AI sycophancy is when a chatbot tells you what you want to hear, agreeing with you or praising your ideas, even when that makes its answer less accurate.
Also called sycophancy
In more detail
Training methods like RLHF reward answers people like, and people tend to like agreement. The result can be a model that backs down when challenged or flatters weak work.
To reduce it, ask for honest criticism, ask the model to argue the other side, or present your idea neutrally instead of as your own.
Example
You insist a wrong answer is right, and the chatbot apologises and agrees, even though its first answer was correct.
Related terms
Reinforcement learning from human feedback (RLHF)Reinforcement learning from human feedback (RLHF) trains an AI model to give better answers by learning from people’s ratings of its responses.AI alignmentAI alignment is the challenge of making an AI system’s goals and behaviour match what people actually intend and value.Automation biasAutomation bias is the tendency to trust a computer’s answer more than we should, and to check it less carefully than we would a person’s.AI hallucinationAn AI hallucination is when an AI confidently gives an answer that is false or made up, such as a fake fact, quote or source.
Browse related tools
AI tools listed on BroUseAI in these categories.
