From the Public Safety edition of November 8, 2026
Federal court declines sanctions for AI-drafted motion with wrong citations but refuses to consider the replacement authorities and warns counsel
U.S. District Court, N.D. California (order summarised by EDRM / Michael D. Berman) · Courts / AI-assisted legal drafting · California, United States
- Publisher
- EDRM
- Original publication
- EDRM, Nov. 3, 2026; order dated Oct. 28, 2026
- Source retrieved
- 2026-11-09
- Event date
- 2026-10-28
What happened
EDRM summarises Farhat v. Cutera, Inc. (N.D. Cal. Oct. 28, 2026). Defendant’s motion to dismiss cited cases the court could not locate; the court issued an order to show cause. Counsel’s declaration said a generative AI drafting tool (Legora) produced the first draft, she checked authorities in Westlaw and corrected errors, but an earlier uncorrected version was converted to PDF and filed. She supplied a redline. The court discharged the order and declined sanctions at this time, but disregarded the inaccurate citations and refused to consider the replacement authorities, and warned that counsel must review the final document and verify authorities exist regardless of AI use or delegation, with future errors risking Rule 11(c) sanctions.
Why it matters
Illustrates a calibrated judicial response: candor and correction avoided sanctions, yet there was still a practical penalty. Relevant to court rules and to prosecutors or public defenders using drafting tools, with limited transfer because this was a civil case.
Evidence and measured results
One case summarised by a practitioner blog; no data on frequency of such errors.
Limitations and uncertainty
Single order; N.D. Cal. practice; civil commercial litigation; not criminal courts or prosecutors; outcome could differ before another judge.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-11-09; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Legal teams adopting drafting tools risk filing unverified or stale text.
- Stakeholders
- Chief prosecutor or public defender
- General counsel
- Court clerk
- Risk manager
- Bar liaison
- Discovery questions
- Which drafting tools are approved and who verifies final filings?
- How do we prove which version was filed?
- What is our protocol after an error is found?
- Value hypothesis
- A verification workflow can reduce the chance of filing unverified authorities; no rate of errors is evidenced.
- Potential engagement
- Pre-filing verification workflow design and training for one legal office.
- Unsupported claims
- That AI errors are widespread in public offices
- That the same leniency would apply elsewhere
- That any tool eliminates fabricated citations
Pre-sales engineering
Role takeaway
- Fit
- Workflow and tooling for verifying the exact filed document.
- Architecture and integration
- Add version locking, citation checking and a tool-use log to the document management and e-filing path.
- Prerequisites
- Document management with version history
- Citation validation service
- Approved tool list
- Constraints
- Validators may miss misquotes
- Court rules vary
- Case data must stay in approved tools
- Security
- Use enterprise-governed AI tools and limit case data exposure.
- Proof of value
- Seed known errors into drafts and check detection in the final PDF
- Measure time added to filing
Delivery
Role takeaway
- Work
- Build and roll out the checklist, logging and training.
- Dependencies
- Tool approval
- Citation checker licence
- Supervisor buy-in
- Ownership
- Head of the legal office with a designated filing verifier.
- Skills and adoption
- Training on Rule 11 duties and on checking the final file.
- Governance checkpoints
- Monthly sample audit of filings
- Error response protocol review
- Annual policy refresh
- Acceptance criteria
- All sampled filings pass authority verification in the final file
- Version record available for every filing
- Risks
- Checklist ignored under deadlines
- Validators give false assurance
- Rules vary by court
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
The failure was a version-control and final-document check problem, not only a model error; filing workflows need verification of the exact file submitted.
Governance
Who approves, reviews and stays accountable for outcomes?
Responsibility is explicitly non-delegable under Rule 11(b); disclosure and redlines were treated as mitigating.
Security and privacy
What data, permissions and controls need testing?
Drafting tools see case material; firms and agencies need approved tools and data handling terms.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
Attorneys keep non-delegable verification duties even when staff or tools prepare drafts; training and supervision are the workforce levers.
Procurement
What should contracts, pricing and exit terms secure?
Procurement should favour enterprise drafting tools with version control and clear data terms; the source does not evaluate any product.
Operating model
Which teams own the service once it runs?
A named verifier and final-file check fit the operating model the court expects.
Publication history
- 2026-11-08Public Safety · Issue 635 resources
Stable resource ID: farhat-v-cutera-ai-citation-errors-no-sanctions-replacement-authorities-disregarded-edrm-20261103