The judge narrative is a plain-English summary that appears at the top of the AI Insights panel on every judge profile. It distills patterns from available tentative rulings, attorney observations, and structured data into a concise description of how this judge approaches their courtroom.
What the narrative synthesizes. The narrative is generated by CaliforniaCourtIntel's AI from three sources: (1) Tentative ruling language — the AI reads the full text of available tentative rulings and identifies patterns in tone, depth of analysis, and reasoning style. (2) Attorney observations — practitioner notes about courtroom demeanor, question style, and procedural preferences are incorporated. (3) Structured data — grant rates, ruling frequency, and appointment history provide context.
The narrative is not a verbatim quote from any single source. It is a synthesis, which means it may describe tendencies that are consistent across many data points without being literally true in every case.
Verifying the narrative. Each narrative includes a View Sources button that shows the specific rulings and observations the AI weighted most heavily. Click through to verify the underlying material before relying on the narrative in professional preparation.
Narrative freshness. Narratives are regenerated when new tentative rulings or observations are added for a judge. The last updated date in the profile header reflects the most recent data refresh. For high-volume judges in major courthouses, this may be as recent as yesterday. For low-volume judges, the narrative may reflect data from several months ago.
When the narrative is limited. If a judge has fewer than 10 data points in our system, the narrative will be short and marked as preliminary. Newly appointed judges or judges in courts without public tentative ruling portals will have limited narrative content. In these cases, the attorney observations section often provides the most useful qualitative insight.
Flagging inaccuracies. If the narrative contains something that does not match your direct experience, click Flag as Inaccurate at the bottom of the AI Insights panel. Flagged narratives are reviewed by our data team within 48 hours.