EN
FR
ES
DE
Factual accuracy assesses whether what an AI asserts about a brand is true and substantiated. An unverifiable or invented claim — however flattering — lowers this score, which stops a fluent hallucination from passing for a good answer.
Speaking well is not enough
An AI can describe a brand in glowing terms while inventing facts: a ranking, a certification, a figure. Factual accuracy (or grounding) checks whether those assertions are true and substantiated, regardless of their tone.
How it is scored
The AI panel tests the assertions against verifiable facts. A specific claim absent from known facts and unverifiable is treated as a hallucination: the accuracy score drops, even if the sentiment is positive.
Why it is decisive
Without this safeguard, a fluent hallucination would pass for an excellent answer. Factual accuracy is the "Quality" dimension of the AGS: it protects the reliability of the score. Detailed example in the demonstration.
Every question asked to ChatGPT without your name in the answer is a competitor recommended instead of you — measured across 6,820 real AI answers.