What is a citation in an AI answer?
A citation is a link from a generated answer to the specific source page or passage the answer was drawn from.
A cited answer tells the reader not just what the answer is but where it came from. In a customer-facing system, that link is usually a page on the business's own website, so following it is both a verification step and a way further into the site.
Why it matters
Citations do two things a merely correct answer does not. For the customer, they make the answer checkable — a claim about a refund window that links to the refund policy can be confirmed in one click, which is the difference between trusting the answer and hoping. For the business, they turn every wrong answer into a repair address: the citation names the page that produced it, so the fix is a specific edit rather than an investigation. Without citations, a complaint about a wrong answer is nearly impossible to act on.
Examples
- 'Standard delivery is three to five working days' linking to the shipping page.
- An answer about plan limits linking to pricing, where a customer can see the whole table.
- A wrong answer whose citation reveals an outdated page nobody had noticed.
Not the same as
Source list
A list of everything retrieved is not a citation. A citation names the passage that produced this answer.
Related
term
What is grounding in AI?
A grounded system retrieves passages before it writes anything, puts them in front of the model, and instructs it to answer only from what it was given. The test of whether a system is grounded is simple and does not require access to its internals: ask it something, then ask it where the answer came from. A grounded system names a page. An ungrounded one describes its reasoning.
term
What is an AI hallucination?
Language models predict plausible text, not true text. Asked a question with no supporting material, a model does not fall silent; it produces the most likely-sounding answer, which is often a wrong price, a policy that does not exist, or a citation to a page that was never written. The behaviour is not a bug to be patched out of the model. It is what the model does when nothing constrains it.
term
What is RAG (retrieval-augmented generation)?
RAG is what separates a chatbot that knows your business from one that guesses. Documents are split into passages and indexed by meaning; each question retrieves the closest passages, and the model is instructed to answer only from them.
term
What is a knowledge gap in customer support?
Every refusal is a record of something customers want to know and the website does not say. Collected over a month, those refusals become a ranked list of what is missing, ordered by how many people asked — which is a content brief nobody had to guess at.
industry
AI customer support for restaurants
Restaurant enquiries arrive at exactly the moment nobody can answer them: during service, on a phone nobody can reach, from someone deciding right now where to eat tonight. Most of what they ask is already on the website. RubyRep reads it and answers instantly, on the site, while the decision is still open.
industry
AI customer support for law firms
People contacting a solicitor are usually anxious and are asking process questions, not legal ones: what this costs, how long it takes, whether the first conversation is confidential. Those answers are on your website and RubyRep can give them instantly — while refusing, by design, to do anything that looks like advice.
RubyRep is this, applied to your website.
Give it your URL and it reads your pages, then answers your customers from them.
Train My RubyRep