Citation
Kutipan
A pointer from a claim back to the material that supports it.
A citation is a link from a specific claim to the specific piece of evidence behind it. Its purpose is checkability: the reader can confirm the claim instead of trusting the speaker.
For AI systems, citations are the practical answer to over-reliance, because they make the chain from question to evidence visible. Without them, a correct answer and a fabricated one look identical from the outside.
They have to be real. A citation that points to a document the system never read is worse than no citation, because it borrows credibility from a source the reader will check and find hollow. Verify that the referenced passage actually contains the claim, which is a small amount of engineering and saves a lot of trust.
Related terms
Grounding
Tying the answer to something checkable, so it can be verified instead of trusted.
RAG
Fetch the relevant documents first, then let the model answer using them.
Over-reliance
Trusting the output because it is well written, not because it is right.
Evaluation
Measuring whether the system does the job, on a fixed set of cases, repeatedly.
Web Search
A tool that lets the model look things up instead of guessing.
Want this working inside your team?
It starts with one conversation about your workflow.