Essandoh v. Capital One: Court Mandates AI Disclosure Certificate for All Filings | Advocate Prakhar

Case at a Glance

Full Citation	Laud Essandoh v. Capital One Bank NA, No. 4:25-CV-00746-ALM-BD, 2026 WL 1040383 (E.D. Tex. Apr. 16, 2026)
Court	U.S. District Court, Eastern District of Texas, Sherman Division
Date	April 16, 2026
Category	AI Disclosure Order & Rule 11
Jurisdiction	United States (Federal, Texas)
AI Tool Used	Unspecified generative AI (admitted by plaintiff)
Outcome/Sanction	Formal admonishment; court-wide AI certification order issued

Background

Laud Essandoh filed suit against Capital One Bank in the Eastern District of Texas. The case was in the discovery phase when the court convened a hearing based on the parties’ joint report, in which Essandoh’s portion contained inaccuracies. At the hearing, Essandoh admitted that these inaccuracies resulted from his reliance on generative AI without ensuring the accuracy of the output. Magistrate Judge Bill Davis admonished Essandoh and issued a supplemental order establishing mandatory AI certification requirements for all filers in the court.

The AI Issue

The central legal question was what procedural safeguards courts must impose when litigants use AI tools to prepare filings and those tools produce inaccurate content. The court also addressed whether Rule 11(b) of the Federal Rules of Civil Procedure is sufficient to govern AI-assisted filings, or whether supplemental local rules and explicit certification requirements are needed to protect judicial integrity.

What the Court Decided

• Any document prepared with AI assistance must include a “Certificate of Generative Artificial Intelligence Usage” disclosing the AI tool used and how it was used [mandatory disclosure rule].
• The filer must certify that they have confirmed the accuracy of every factual, procedural, or legal assertion produced with AI assistance [personal responsibility principle].
• The court will presume any filing without such a certificate was prepared without AI [rebuttable presumption rule].
• Attorneys and pro se litigants who sign AI-assisted filings are held personally responsible for the content under Rule 11(b) [attorney liability rule].
• AI-generated inaccuracies in discovery documents are treated as seriously as inaccurate citations in briefs [broad application of Rule 11].

“The opposing party wastes time and money in exposing the deception. The Court’s time is taken from other important endeavors. The client may be deprived of arguments based on authentic judicial precedents.”

— Citing Mata v. Avianca, Inc., 678 F. Supp. 3d 443, 448-49 (S.D.N.Y. 2023), applied by Judge Davis, April 16, 2026

The India Angle

Indian Law Equivalent

India has no equivalent court-wide AI certification rule yet, but the framework is emerging. The Advocates Act, 1961 (Section 35) and the Supreme Court Rules, 2013 already require candour and accuracy in all court submissions. The Bar Council of India’s Rules of Professional Conduct operate alongside Order XI CPC, which governs discovery and document production. AI-generated inaccuracies in affidavits or discovery documents could expose parties to proceedings under Order XI Rule 21 (contempt for false affidavits) or Section 193 IPC (false evidence), now replicated in the Bharatiya Nyaya Sanhita 2023 (BNS), Section 229.

Bar Council Rules

BCI Rule 14 (no false statements to court), Rule 15 (no conduct bringing disrepute), and Rule 36 (professional standards in correspondence and submissions) are all engaged when AI-generated content is filed without verification. The Essandoh-style certification order, if adopted by Indian High Courts or the Supreme Court, would sit naturally within the existing framework of professional responsibility under the Advocates Act.

Practical Advice for Indian Advocates

• Voluntarily adopt a personal AI certification practice now: note in your affidavit or covering letter when AI tools were used in drafting and confirm all citations were independently verified.
• When preparing discovery documents, affidavits, or joint reports with AI assistance, require a senior advocate or partner to personally review and verify all factual assertions before signing.
• Monitor Bar Council of India circulars and Supreme Court Practice Directions for forthcoming AI disclosure requirements, which are likely to follow the global trend established in Essandoh.

Quick Takeaways

• Disclose AI use in filings and certify accuracy of every AI-generated assertion.
• Personal responsibility under Rule 11 extends fully to AI-assisted court documents.
• Courts presume non-AI drafting where no certificate is provided — get it in writing.

---

Deep Dive: The Three-Part AI Certification Framework and Its Global Implications

The Essandoh order is one of the most practically significant AI-related court orders of 2026 because it moves beyond punishment and into prevention. Rather than waiting for AI hallucinations to infect court filings and then sanctioning filers, Magistrate Judge Davis constructed a proactive three-part framework: mandatory disclosure, named-tool identification, and filer verification. This approach directly addresses the root cause of AI hallucination problems in courts — not the technology itself, but the absence of human accountability for AI-generated output.

The catalyst was straightforward: Essandoh, a pro se plaintiff, admitted at a discovery hearing that inaccuracies in a joint report reflected unchecked AI reliance. This was not a case of sophisticated misconduct; it was a case of misplaced trust in technology that courts are now required to address systemically. The Eastern District of Texas’s supplemental order explicitly builds on and reinforces existing Local Rules AT-3(m) and CV-11(g), demonstrating that courts are layering AI-specific requirements on top of existing professional conduct rules rather than treating AI governance as an entirely separate area of law.

Compared to Mata v. Avianca, where retrospective sanctions were imposed after discovery of fabricated citations in an already-filed brief, the Essandoh approach is preventive architecture. Several courts in 2025 and 2026 have adopted similar frameworks, including the Northern District of California and various Southern District of New York judges. The direction of travel is clear: AI disclosure and personal filer verification are becoming standard elements of federal court practice in the United States.

For Indian advocates, the Essandoh framework offers a model for voluntary best practice in advance of formal regulatory adoption. India’s court system processes enormous volumes of filings, and the temptation to use AI for research and drafting — particularly in high-volume commercial, insolvency, and consumer dispute tribunals — is significant. An Essandoh-style certification protocol, voluntarily adopted by progressive law firms and bar associations, would both protect advocates from professional liability and build judicial confidence in AI-assisted practice. The question is not whether Indian courts will eventually require AI disclosure, but when — and early adopters will be ahead of that curve.