For each source URL, the Veracity Engine determines how accurately an AI model represented the source by calculating a fidelity score based on three categorized fact sets:

\[\text{Score}_{\text{URL}} = \frac{|\text{survived}|}{|\text{survived}| + |\text{lost}| + |\text{distortions}|} \times 100\]

In the edge case where no relevant facts are detected (\(|\text{total}| = 0\)), the URL receives a default score of \(100.0\%\). Once all individual source assessments are complete, the run's overall fidelity score is derived by computing the arithmetic mean across all valid, non-null URL scores rounded to one decimal place.