Lnu v. Blanche
U.S. Court of Appeals for the Ninth Circuit · 9th Cir. · Alaska bar guidance , Arizona bar guidance , California bar guidance , Hawaii bar guidance , Idaho bar guidance , Montana bar guidance , Nevada bar guidance , Oregon bar guidance , Washington bar guidance
Conduct
Firm had unlicensed graduates draft immigration briefs with AI; no attorney checked the citations, and counsel denied AI use at argument.
Consequence
$2,500 each, six-month Ninth Circuit suspension for both attorneys, two-year sworn AI disclosure on all firm filings, State Bar referral.
Lesson
Delegating drafting does not delegate the signature: the signing lawyer answers for every citation, and denial compounds the sanction.
Verified September 28, 2026
- Citation
- Lnu v. Blanche, 177 F.4th 1014 (9th Cir. 2026) (published order), No. 24-4790 (June 3, 2026)
- Decided
- June 3, 2026
Summary
In an immigration petition for review, attorney Mike Singh Sethi of Sethi Law Group/US Legal Group, APC filed an opening brief that cited two nonexistent Ninth Circuit decisions ("Eduardo v. Garland, 28 F.4th 742" and "Lay v. Holder, 729 F.3d 962") and attributed quotations to two real opinions in which the quoted language does not appear; the reply brief misstated three more. The court found similar fabrications in briefs Sethi filed in two other petitions. The firm had unlicensed law-school graduates ("Brief Writers") write the legal arguments, and Sethi and Rounds explained that they "do not normally vet citations used by the Brief writer during [their] review." At oral argument Rounds said AI "was not used"; after an order to show cause, Sethi and Rounds conceded it was "more likely than not" or "probable" that the errors came from unauthorized AI use by the Brief Writers and a failure to cite-check. Their first response to the problem, a motion to correct that described the two nonexistent cases as "typographical errors" and swapped in replacement cases, also cited cases that did not support the propositions.
- AI tool:
- Generative AI (counsel conceded AI use by the firm's unlicensed brief writers was "more likely than not"; tool not identified)
- Sanction amount:
- $5,000 total ($2,500 each against attorneys Mike Singh Sethi and William Rounds); six-month suspension of both from practice before the Ninth Circuit; notice to clients, courts, and firm attorneys; two-year sworn AI-disclosure certification on all future filings by every attorney at the firm; referral to the State Bar of California
What sanction did the court impose?
In a published order, the panel disciplined both attorneys under Fed. R. App. P. 46(b) and Ninth Circuit Rule 46-2. Sethi and Rounds were each sanctioned $2,500, payable within 21 days, and suspended from practice before the Ninth Circuit for six months starting ten days after the order. They must give a copy of the order to their clients, opposing counsel, and the presiding judge in every pending case and to every attorney in their firm. For two years, Sethi, Rounds, and all attorneys at the firm must include in all future filings a statement under penalty of perjury addressing whether generative AI was used, naming the tool, and certifying personal review and that all citations and quotations refer to existing authority. The Clerk was directed to serve the order on the State Bar of California. The underlying petition had been granted in November 2025.
Why does Lnu v. Blanche matter for law firms using AI?
Lnu v. Blanche is a published disciplinary order, and its reasoning is structured to outlast any particular technology. The panel was explicit: “We do not sanction Sethi and Rounds for the simple fact that they or their subordinates used generative AI,” and it is “ultimately irrelevant to the disciplinary analysis (except for the duty of candor), however, whether Sethi, Rounds, or anyone at the Firm actually used generative AI … .” The violations fall into two groups. Sethi, whose name alone appeared on the briefs, violated Fed. R. App. P. 28(a)(8)(A) and California Rules of Professional Conduct 1.1, 1.3, and 3.1(a)(2) when he signed and filed them. Sethi and Rounds both violated the duty of candor under Rule 3.3(a)(1) when they represented the errors as innocent typographical mistakes and denied that generative AI might have been the source.
Candor drove the severity. The first corrective filing called the nonexistent cases “typographical errors” and replaced them with others that also failed to support the propositions. At oral argument Rounds said AI “was not used,” then conceded it was “possible.” The court said that had the attorneys disclosed the AI use and apologized in the motion to correct, “lesser sanctions may have been warranted.” The court used the order to state a disclosure rule: an attorney who discovers a hallucination in a filed brief must notify the court and opposing counsel immediately, describe the nature of the error, and disclose how it came about. “The gravity of discipline we impose, including the temporary suspension of practice, is owed to this repeated failure of candor.”
The remedies reach beyond the two lawyers. The notice requirement puts the order in front of every client and every presiding judge in their pending cases, and the two-year certification binds every attorney at the firm in “all future filings,” requiring a sworn statement on whether generative AI was used, the name of the tool, personal review of the filing, and that all citations and quotations refer to existing authority. For a firm that relies on non-attorney drafters, the order is a direct statement of what the Ninth Circuit expects the signing attorney to have done before the brief is filed.
Implications for your firm
Operational steps a firm reading this case may wish to consider documenting. Strategic and rule-application calls belong to your firm's attorneys.
- Review how briefs drafted by nonlawyers or junior staff are checked before signature. The court called the firm's admission that attorneys did not vet the Brief Writers' citations 'an extraordinary confession' and flagged possible issues under California Rules of Professional Conduct 5.3 and 5.5.
- Document the correction protocol for a discovered hallucination. The panel held that, in circumstances like these, counsel must notify the court and opposing counsel immediately, describe the nature of the error, and disclose how it came about; a bare errata swap is not sufficient disclosure.
- Train attorneys that candor failures drive the severity of discipline. The court tied the suspension to the attorneys' 'repeated failure of candor,' not to AI use as such.
- Consider that a firm policy prohibiting AI does not answer the verification duty. The attorneys said they did not suspect AI because the firm prohibited it; the panel disciplined Sethi for signing and filing unverified briefs, and both attorneys for their later statements about the errors.