Oneto v. Watson
U.S. District Court, Northern District of California · N.D. Cal. · California bar guidance
Verified May 5, 2026
- Citation
- Oneto v. Watson, No. 22-cv-05206-AMO, Dkt. No. 92 (N.D. Cal. Oct. 10, 2025)
- Decided
- October 10, 2025
Summary
Plaintiff's counsel Edward A. Quesada submitted a Rule 52 brief citing three nonexistent cases (Ziskind v. Spector, Moffitt v. Whiting-Turner, and Chandhok v. Companion Life Ins. Co.) with fabricated Westlaw citations. Quesada acknowledged that he "may have unknowingly copied and pasted material from an AI overview" found through Google, and stated he had been unaware Google had an AI feature. Judge Araceli Martínez-Olguín found Quesada violated Federal Rule of Civil Procedure 11(b) and California Rules of Professional Conduct 3.1 and 3.3.
- AI tool:
- Google AI Overview
- Sanction amount:
- $1,000
What sanction did the court impose?
$1,000 monetary sanction payable to the Clerk of Court, mandatory service of the order on the client, completion of at least one hour of CLE on the ethical use of AI in legal practice within 60 days, and referral of the order to the State Bar of California.
Why does Oneto v. Watson matter for law firms using AI?
Oneto is a notable variant of the AI hallucination pattern: the sanctioned attorney denied knowingly using generative AI at all, attributing the fictitious citations to Google’s AI Overview surfacing in ordinary search results. Judge Martínez-Olguín’s order makes clear that the source of the fabricated authority is irrelevant to Rule 11 liability, and that failure to recognize AI in routine research tools may itself signal a Rule 1.1 competence problem. Firms relying on attorneys who treat search engines as neutral retrieval should treat this as a direct warning.
Sources
Primary sources
Further reading
Source PDF is a Westlaw printout mirrored from the Damien Charlotin hallucination database. We are working to add the underlying court docket (PACER, CourtListener, or court website) as a second source.