Morgan v. V2X Inc.: Court Orders Disclosure of AI Tools Used with Confidential Information | Advocate Prakhar

Case at a Glance

Full Citation	Morgan v. V2X, Inc., Civil Action No. 25-cv-01991-SKC-MDB, D. Colo. (March 30, 2026), Doc. No. 65
Court	U.S. District Court, District of Colorado, Magistrate Judge Maritza Dominguez Braswell
Date	30 March 2026
Category	AI Confidentiality, Protective Orders & Discovery
Jurisdiction	USA — Colorado (Federal)
AI Tool Used	Undisclosed (mainstream AI tool such as ChatGPT, Claude, or Gemini suspected)
Outcome	Motion granted in part; protective order amended to prohibit uploading confidential info to mainstream AI; plaintiff ordered to disclose AI tool used within 10 days

Background

Archie Morgan, a self-described “qualified Black American manager,” sued his former employer V2X, Inc. for racial discrimination, hostile work environment, and wrongful termination in an employment discrimination case pending in the District of Colorado. Morgan represented himself pro se. Defendant V2X moved to amend the stipulated protective order to include AI-specific language, arguing that Morgan may have uploaded confidential discovery materials to a mainstream AI platform. The dispute arose after Morgan moved to compel production of an insurance policy, and V2X counter-moved for AI disclosure and use restrictions as a precondition.

The AI Issue

The core legal question was twofold: first, whether Federal Rule of Civil Procedure 26(b)(3)’s work-product protection shields a pro se litigant from disclosing which AI tool they used; and second, whether a protective order can prohibit uploading confidential materials to commercial AI platforms. The court held that identifying the AI tool’s name — without revealing prompts or analytical outputs — does not expose mental impressions or litigation strategy, and that the defendant had a legitimate interest in determining whether confidential information had been compromised through AI data-retention or training practices.

What the Court Decided

• Work-product protection under FRCP 26(b)(3) does not shield the identity of an AI tool used with confidential materials — only mental impressions and litigation strategy are protected [work-product doctrine scope].
• Protective orders can be amended to include AI-specific provisions prohibiting upload of confidential information to mainstream AI platforms like standard ChatGPT, Claude, or Gemini [protective order amendment].
• Plaintiff was ordered to disclose the name of the AI platform used to process any confidential discovery materials within 10 days [disclosure obligation].
• Defendant was then ordered to file a Notice within 10 days of disclosure indicating whether it will seek relief or has mitigated the potential harm [follow-up procedure].
• The court noted a growing AI equity gap between large firms with enterprise-grade AI and pro se litigants using free tools — but this gap did not relieve the plaintiff of compliance obligations [access-to-justice observation].

“This highlights a growing problem in the age of AI: as large firms pour thousands of dollars into enterprise-grade AI and make their use of AI more secure, efficient, effective, and powerful, how will a pro se litigant or a litigant who cannot afford big-ticket legal services and better AI keep up?”

— Magistrate Judge Maritza Dominguez Braswell, D. Colorado, 30 March 2026

The India Angle

Indian Law Equivalent

In India, confidential discovery materials are governed by the Code of Civil Procedure, 1908 (Order XI) and protective orders issued by civil courts under their inherent powers (Section 151 CPC). The Information Technology Act, 2000 (Section 43A) and the Digital Personal Data Protection Act, 2023 impose obligations on entities handling personal data — uploading confidential litigation materials to a foreign AI platform may constitute a cross-border data transfer requiring consent or exemption under the DPDP Act. Indian courts have inherent power to restrict use of AI tools with confidential court materials.

Bar Council Rules

Under Bar Council of India Rules, Part VI, Chapter II, Section II, Rule 17, advocates are bound by client confidentiality. Uploading a client’s confidential litigation materials to a commercial AI tool whose training data retention policies are unclear could constitute a breach of the advocate’s duty of confidentiality. The BCI has not yet issued specific AI guidance, making awareness of cases like Morgan v. V2X critical for Indian practitioners managing discovery materials.

Practical Advice for Indian Advocates

• Never upload client-confidential documents or court-designated protected materials to consumer-grade AI tools — use only air-gapped, enterprise-grade, or locally-hosted AI models with explicit no-training-retention guarantees.
• When dealing with discovery materials or documents covered by protective orders in Indian civil proceedings, treat AI tools as a third-party disclosure risk and seek explicit client consent before using AI for document analysis.
• Be aware that Indian courts may soon follow US precedent and require disclosure of AI tools used in processing discovery — maintaining an AI-use log in your litigation files is advisable practice now.

Quick Takeaways

• The identity of an AI tool used with confidential discovery materials is not protected by work-product privilege.
• Protective orders can and will be amended to explicitly restrict mainstream AI tool use with confidential litigation data.
• The AI equity gap between large firms and pro se litigants is a growing judicial concern but does not excuse non-compliance.

---

Deep Dive: AI Confidentiality as the Next Frontier in Litigation Ethics

Morgan v. V2X breaks new ground not in the hallucination space but in the equally urgent domain of AI confidentiality in active litigation. Unlike the wave of 2026 cases involving fake citations, this case grapples with a subtler problem: what happens when a litigant uploads real, confidential, court-designated materials to a commercial AI platform that may retain, process, or learn from those materials? The court’s decision that disclosure of the AI tool’s name does not implicate work-product protection is a logical extension of existing doctrine — identifying a research database (Westlaw, LexisNexis) has never been considered work-product — but applying it to AI is novel and consequential.

The court’s acknowledgment of the AI equity gap is one of the most thoughtful passages in 2026 AI-in-courts jurisprudence. Large law firms are deploying Microsoft Copilot for Law, Harvey AI, and custom enterprise LLMs with contractual no-training guarantees and security certifications. A pro se plaintiff like Morgan has access to free-tier ChatGPT, which until recently used conversation data for model improvement. This asymmetry creates a structural disadvantage: the corporate litigant’s AI use is presumptively safe; the individual’s is presumptively risky. The court did not resolve this equity problem — it could not — but naming it publicly creates pressure on courts, bar associations, and legal aid organizations to develop AI-access frameworks that do not inadvertently penalise indigent litigants for using the only tools available to them.

For practitioners in India, the Morgan ruling arrives at a critical moment. India’s Digital Personal Data Protection Act, 2023 introduces data principal rights and cross-border transfer restrictions that intersect directly with the scenario in this case. An Indian advocate uploading a client’s confidential affidavit or commercial contract to a US-based AI platform may be triggering DPDP Act compliance obligations without realising it. The Act’s provisions on “significant data fiduciaries” and consent requirements for cross-border transfers are particularly relevant. Indian High Courts with commercial divisions may soon encounter disputes about whether uploading discovery materials to AI tools constitutes a breach of protective orders or statutory confidentiality duties.

The Morgan case also signals a coming wave of AI-disclosure obligations in US federal litigation. Several district courts have already adopted standing orders requiring disclosure of AI tool use in filed documents. The protective order amendment in Morgan extends this transparency principle to the discovery process itself — not just the documents filed with the court. Indian litigants and their counsel who appear before US courts, or who deal with cross-border discovery requests, should take specific note: AI tool disclosure is now a live compliance issue, not a theoretical future concern.