By Kim Xi Harris | Founder & Platform Architect, Lex Arca™ Legal Vault | Calculate Your Firm’s Billing Leakage | legalvault@lex-arca.com
According to Clio’s 2026 Legal Trends Report for Solo and Small Law Firms (May 2026, https://www.clio.com/about/press/2026-solo-small-firm-report/), 71% of solo practitioners and 75% of small firms are now using AI to complete legal work — yet fewer than 33% have seen any revenue increase from it, compared to nearly 60% of enterprise firms. The gap between AI adoption and AI compliance is not a policy problem. It is an architecture problem.
In May 2026, two managing partners at Quinn Patton — a Severna Park, Maryland law firm — were ordered by U.S. District Judge Stephanie Gallagher to explain ‘rampant citation issues’ across multiple federal filings after AI hallucinations appeared in at least five briefs. One partner had already been warned by a D.C. federal judge in October 2025 that further violations ‘will not be tolerated and may result in sanctions and/or referral to a disciplinary committee.’ He filed another hallucinated brief within the month. The lesson for Maryland litigators is not that AI is dangerous. It is that AI without a documented verification workflow is professionally indefensible.
What Happened in the Quinn Patton Cases — and Why It Matters Beyond Maryland?
Donald Quinn and Katherine Patton, both admitted to the Maryland and D.C. bars in 2021, were the managing partners of Quinn Patton LLC in Severna Park. According to court records reviewed by the Maryland Daily Record, the firm had been using a closed AI system to edit legal briefs — one described by Quinn as relying only on cases researched through Westlaw. The problem: the AI tool converted standard case citations into fabricated quotations during the editing process.
The result was five filings across multiple federal courts containing case citations that either did not exist or fundamentally misrepresented the law. Three of those cases were before the same judge — U.S. District Judge Stephanie Gallagher in the District of Maryland — who issued separate orders in each case requiring both attorneys to explain the ‘hallucinations.’ At the U.S. District Court for the District of Columbia, Judge Rudolph Contreras had already put Quinn on notice in October 2025, writing in a memorandum opinion that Quinn had ‘apparently violated’ both the Federal Rules of Civil Procedure and D.C. Bar rules of professional conduct.
The pattern that emerged was not a one-time error. It was a systemic absence of any verification step between AI output and court filing. Less than a month after Contreras’s warning, the firm filed another brief with hallucinated citations.
‘The reality is, this is only going to happen more and more. Unless the courts start leveling some real consequences against attorneys, there’s no incentive for the attorneys to slow down.’ — James Rubinowitz, attorney and AI law lecturer, Cardozo School of Law
What Does ‘Closed System’ Actually Mean — and Why Wasn’t It Enough?
Quinn’s description of the firm’s AI setup is worth examining carefully, because it illustrates a misconception that likely affects many small litigation firms right now. He described using a ‘closed system’ — meaning the AI did not share data outside the firm. This is a data privacy control. It is not a compliance architecture.
A closed system that restricts data egress does not verify AI output against actual case law. It does not run a jurisdictional check before synthesis. It does not produce a Verification Attestation documenting what the AI generated, against what parameters, and when. And it does not create an append-only, tamper-evident audit trail that demonstrates attorney review occurred before the filing went out.
The Quinn Patton situation is the clearest possible illustration of the gap between ‘our data stays inside the firm’ and ‘we have a documented, verifiable AI compliance workflow.’ The first is a vendor contract feature. The second is what ABA Formal Opinion 512 actually requires — and what courts are now enforcing.
Under ABA Formal Opinion 512, every licensed attorney must maintain a reasonable understanding of any AI tool used in their practice, personally verify all AI-generated output before filing, and disclose AI use to clients when it is material to the representation. This obligation is enforceable under Model Rules 1.1, 1.4, and 1.5. It cannot be satisfied by pointing to a closed system. It requires documented evidence that the attorney reviewed and verified the output.
What Are Maryland and D.C. Federal Courts Now Requiring of Attorneys Using AI?
The Quinn Patton cases add Maryland’s federal district courts to a growing national enforcement map. As of May 2026, more than 300 standing court orders govern AI use in filings across the United States — a number that grew by more than 200 in the second half of 2025 alone. Florida’s Administrative Order 26-04, issued in January 2026, requires personal attorney certification on every AI-assisted filing. Texas requires attorneys to personally certify they reviewed every AI-assisted statement.
In the District of Maryland and the District of Columbia, enforcement is now operating through individual judicial orders rather than a district-wide rule — which means the obligation falls on the attorney to know whether the judge before whom they are appearing has issued standing AI guidance, and to meet that standard without being explicitly told to. Judge Gallagher’s orders in the Quinn Patton cases signal that Maryland’s federal bench is not waiting for a formal district rule before acting.
Judge Gallagher wrote that her concerns ‘extend to counsel’s entire firm’ — and made clear that she would not accept a partially corrected brief, writing that ‘attorneys are not entitled to a second bite at the apple’ after filing briefs containing hallucinated citations and quotations. That is the standard Maryland litigators are now operating under, whether or not a formal order has been issued in their specific courtroom.
What Verification Workflow Would Have Prevented the Quinn Patton Outcome?
The Quinn Patton cases were not a technology failure. They were a process failure. The firm’s AI tool produced errors during editing. Those errors reached filed briefs because no step in the workflow was designed to catch them before filing. A compliance-grade AI workflow for litigation attorneys includes four steps that the firm’s closed editing system did not have.
First, a Sentinel jurisdictional gate checks active court orders and bar rules for the filing jurisdiction before any AI synthesis runs — so the attorney knows, before the AI generates output, whether the target court has standing restrictions on AI use. Second, the AI synthesis runs, bounded by those jurisdictional parameters. Third, a Verification Attestation certificate is generated — an ABA Formal Opinion 512-aligned document that records what the AI produced, under what jurisdictional constraints, and when. Fourth, the certificate and the AI activity are cryptographically timestamped in both the activity log and the billing ledger, creating an append-only, tamper-evident documented activity trail.
This is not a theoretical workflow. It is the AI compliance certification workflow that solo and small-firm litigators need to operate under the enforcement standard now visible in Maryland, D.C., Florida, and Texas federal courts. An ABA Opinion 512 compliance workflow built into the attorney’s litigation vault means the documentation exists before the filing goes out — not as a defense after the sanction arrives.
What Should Maryland Solo and Small-Firm Litigators Do Right Now?
The practical response to the Quinn Patton cases does not require abandoning AI tools. It requires building the verification layer that was missing from the firm’s workflow. For the solo or small-firm litigator in Maryland or the D.C. circuit, that means three immediate actions.
Review every AI tool currently in use against the four-step compliance framework above. If the tool does not generate a Verification Attestation and does not produce a documented activity trail, it is not compliant with ABA Formal Opinion 512 regardless of how its data privacy controls are described. A closed system is not a compliant system.
Establish a written AI use policy before the next filing. Judge Gallagher’s concern extending to ‘counsel’s entire firm’ means the obligation is firm-wide, not attorney-by-attorney. A written policy that requires human review and verification of all AI-generated output before filing is the minimum floor.
Document attorney review explicitly, every time. The emerging judicial standard — visible in Gallagher’s orders, Contreras’s warning, and Florida Administrative Order 26-04 — is that the attorney must be able to demonstrate, with documentation, that they personally verified AI-generated content before it went to the court. That documentation must exist in a form the court can review. A litigation intelligence platform for solo firms that generates a Verification Attestation certificate on every AI-assisted research session creates that documentation automatically.
Key Takeaways
- In May 2026, two Maryland federal court judges ordered Quinn Patton attorneys to explain AI hallucinations in at least five filed briefs, following an October 2025 warning from a D.C. federal judge — establishing a documented enforcement pattern across both districts.
- ABA Formal Opinion 512 requires personal attorney verification of all AI-generated output, enforceable under Model Rules 1.1, 1.4, and 1.5 — a standard that a ‘closed’ AI system with no verification workflow does not satisfy.
- Maryland and D.C. federal litigators must assume that individual judges are enforcing AI standards through case-specific orders even where no district-wide rule has been issued — and must maintain documented verification records accordingly.
- Lex Arca Legal Vault provides a documented, verifiable AI activity trail designed to support attorney compliance workflows — including a Sentinel jurisdictional gate, Verification Attestation certificate, and cryptographically timestamped audit trail built for solo and small-firm litigators.
- Calculate your firm’s billing leakage and get early access at https://calculator.lex-arca.com.
About the Author: Kim Xi Harris is the Founder and Platform Architect of Lex Arca™ Legal Vault, an AI-native litigation intelligence and compliance platform for solo and small-firm attorneys. She is a Cornell Women’s Entrepreneur Program graduate, SBA Women in Business Champion Award recipient, WOSB certified, and holds five Google AI certifications. Calculate your firm’s billing leakage and join the VIP waitlist at https://calculator.lex-arca.com — or reach us at legalvault@lex-arca.com.
