June 1, 2026 (in 3 days): New York: 22 NYCRR Part 161 takes effect, system-wide AI policy for all UCS courts

Williams v. Honl

Court of Appeals of Oregon · Or. Ct. App. · Oregon bar guidance

Conduct

Six of nine cases in an opening brief had fabricated quotes; counsel moved to amend without disclosing AI use.

Consequence

Brief struck, $8,044 in opposing fees awarded; firm jointly liable. Future briefs require AI-free certification.

Lesson

Fabrications that 'permeate' a brief are not salvageable by amendment. Oregon now imposes fees plus certification.

Court sanction

Verified May 5, 2026

Citation
Williams v. Honl, 348 Or. App. 505 (Or. Ct. App. Apr. 22, 2026) (No. A186656)
Decided
April 22, 2026

Summary

Appellant's attorney Abby Shearer submitted an opening brief in which six of nine cases were cited for nonexistent quotations or for propositions of law not plausibly attributable to them, including a fabricated quotation purportedly from Tubra v. Cooke, 233 Or App 339 (2010). After respondent's answering brief flagged the issues, Shearer moved to file an amended opening brief without disclosing the AI-generated fabrications. Chief Judge Lagesen, writing for a panel that also included Presiding Judge Egan and Judge Joyce, struck the brief and ordered Shearer to show cause. Shearer then acknowledged that generative artificial intelligence had been used as a research and drafting aid and that quotations and propositions had not been independently verified.

AI tool:
Unspecified generative AI
Sanction amount:
$8,044.25
This case summary is informational only. Verify the underlying opinion or order against the primary source before relying on it in any filing or client matter.

What sanction did the court impose?

The Court of Appeals struck the opening brief, denied the motion to amend, and ordered appellant's attorneys to pay $8,044.25 in respondent's attorney fees under ORAP 1.40 and ORCP 17, with the law firm jointly liable under ORCP 17 D(1). Appellant was permitted to file an amended opening brief within 28 days; if filed by the same attorney, it must include a certification that counsel drafted the brief without generative AI, has read each cited source, and has verified that every cited source exists. The court declined to dismiss the appeal or to impose additional sanctions payable to the court.

Why does Williams v. Honl matter for law firms using AI?

Williams v. Honl is the first Oregon appellate decision to award attorney fees, rather than only court-payable sanctions, against counsel responsible for AI-generated fabricated authority, and it joins a growing line of Oregon Court of Appeals cases (Ringo, Powell, Doiban) building a fee-shifting and certification regime around generative AI misuse. For managing partners, the operative warning is structural: the brief was not salvageable because the fabrications “permeated” it, and the court held that “cite-checking” AI output bears no resemblance to the competent practice of law. Firms whose appellate practice touches Oregon should expect a Ringo-style certification to become standard whenever AI use is even suspected.

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.