Jump to content

Gillespie v Superior Court

From AI Law Wiki

AI-generated text. This page was generated using artificial intelligence.

Case Information
Case name Crystal Gillespie v. Superior Court of San Diego County; The People, real party in interest[1]
Court California Court of Appeal, Fourth Appellate District, Division One[1]
Docket D088222 (writ proceeding); SCD307508 (underlying superior-court case)[1]
Filed May 2026 (writ petition; exact day not identified in the reviewed opinion)[1]
Judge(s) Do, J.; McConnell, P. J.; Rubin, J. (appellate panel); Melinda Lasater (trial judge)[1]
Petitioner Crystal Gillespie[1]
Respondent / real party Superior Court of San Diego County (respondent); The People (real party in interest)[1]
Case type Original writ proceeding concerning criminal pretrial mental health diversion[1]
Claims / issues Independence and reliability of ChatGPT-assisted expert testimony; diversion eligibility under Penal Code § 1001.36; Evidence Code §§ 801–802 and Kelly/Frye reliability requirements[1]
Status November 4, 2026: petition for writ of mandate denied; denial of pretrial mental health diversion left in place. The opinion is unpublished. Later proceedings have not been verified.[1]

Gillespie v. Superior Court is an unpublished California appellate decision concerning ChatGPT-assisted forensic psychology testimony in a criminal pretrial mental health diversion proceeding. On November 4, 2026, the Court of Appeal denied Crystal Gillespie’s petition for a writ of mandate challenging the refusal of diversion.[1]

Proceeding and disputed expert evidence

Gillespie, facing a grand-theft charge, sought pretrial mental health diversion in January 2026. The parties agreed that she had qualifying diagnoses but disputed whether her disorders were a significant factor in the alleged offense. After a March hearing, the trial court found her ineligible for diversion. It did not reach the separate suitability criteria.[1]

The prosecution’s psychologist disclosed using AI to help prepare her report. She testified that she used ChatGPT to organise information, polish language and act as a sounding board, but independently formed her conclusions. The defence challenged that account and the report’s reliability, including an inaccurate statement about the medical record. The trial judge credited the psychologist’s explanation that AI had not determined her opinion.[1]

November 4, 2026 decision

The appellate court expressed concern about the reported AI use and the report’s factual error but deferred to the trial court’s supported credibility finding. It found no abuse of discretion in admitting the expert opinion on that record.[1]

The panel distinguished organisational assistance from using AI to generate a forensic psychologist’s expert opinion. It stated that the latter would constitute a novel scientific technique requiring an adequate foundation for reliability, including general acceptance in the relevant scientific community. Here, the trial court had accepted testimony that the expert independently analysed the evidence and reached her own conclusions.[1]

The court also found sufficient evidence to uphold the determination that the prosecution had rebutted the statutory presumption linking the qualifying disorders to the charged offense. It relied on the expert’s assessment of surveillance footage and testimony, rather than treating an absence of corroboration alone as enough, and denied the writ petition.[1]

Limits of the decision

The opinion was not certified for publication and carries the citation restrictions identified in California Rules of Court, rule 8.1115. It resolves this writ challenge rather than the underlying criminal charge.[1]

A separate objection that providing medical information to the AI platform violated privacy rights was not renewed in the writ petition, so the appellate court did not decide it. The psychologist’s description of a business associate agreement was testimony, rather than a judicial determination of privacy compliance.[1]

References