Skip to content

Generative AI & LLMs

Sycophancy

AI agreeing with or flattering someone instead of giving a more accurate or sensible answer.

Example

An AI system endorses a mistaken claim after the user states it confidently.

Why people use it

It names agreement that makes AI seem supportive while weakening honest feedback.

What you'll hear

“It keeps agreeing even when I change my opinion.”

What this means for you

Ask for evidence and contrary considerations when testing an idea.

Can you control it?

No

No direct control. This describes a wider issue, concept or result rather than something you can simply switch on or off in a tool.

Common questions

Does agreement provide independent confirmation?
No. The answer may reflect conversational pressure rather than evidence.
Can asking a leading question influence the reply?
Yes. Wording that strongly suggests a preferred answer can encourage agreement.
Is polite disagreement possible?
Yes. A helpful system can challenge a claim without being rude or dismissive.

Related terms

Still have questions?

Up to 500 characters.

Ask LATHIC about AI. Relevant glossary entries may be included.

Your question, the glossary entries it matches, and a rotating pseudonymous identifier go to Microsoft Azure’s OpenAI service through Vercel AI Gateway to generate an answer. Zero retention and no training are required of the provider, and LATHIC does not save your question or answer. Privacy Notice