Category: Entities & Citations
What Is Citation Accuracy?
Citation Accuracy is the degree to which an AI citation correctly represents the information contained in the cited source.
An AI system can cite a legitimate source but still misunderstand, oversimplify, or incorrectly attribute the information.
For AI Visibility, citation accuracy is therefore different from simply receiving a citation.
Why Citation Accuracy Matters
A citation creates an association between an AI-generated answer and a source.
If the information is represented incorrectly, the citation can contribute to:
- Incorrect brand representation
- Outdated product information
- Misleading recommendations
- Incorrect company descriptions
- Confusion between entities
- Loss of user trust
Accurate citations help ensure that AI systems use sources in the way those sources actually support.
Example
A company’s product page states:
The platform supports teams of up to 50 users on the standard plan.
An AI answer says:
The platform supports unlimited users on its standard plan.
The company page may be correctly cited, but the AI-generated claim is inaccurate.
This is a Citation Accuracy problem.
Citation Accuracy vs Citation Relevance
These concepts measure different things.
Citation Relevance asks:
Does this source actually fit the question or claim?
Citation Accuracy asks:
Does the AI-generated statement accurately reflect what the source says?
A citation can therefore be:
- Relevant and accurate
- Relevant but inaccurate
- Irrelevant but accurate
- Irrelevant and inaccurate
The strongest citation combines relevance and accuracy.
Citation Accuracy vs Citation Quality
Citation Quality evaluates the overall usefulness and strength of a cited source.
Citation Accuracy focuses specifically on whether the AI’s use or representation of that source is factually correct.
A highly authoritative source can still be inaccurately interpreted.
What Can Go Wrong?
Citation Accuracy problems can occur when AI systems:
- Misread a statement
- Remove an important qualification
- Use outdated information
- Combine facts from different pages
- Attribute information to the wrong company
- Confuse similar products
- Interpret a limitation as a capability
- Generalize a specific claim
- Present an example as a universal fact
Why Context Matters
Information often depends on conditions.
For example:
“Available to customers in Europe.”
may actually mean:
“Available to custome
