Artificial intelligence (“AI”) tools like ChatGPT, Claude, and Gemini have become go-to resources for busy executives trying to organize their thoughts, research legal concepts, and prepare talking points before meeting with their attorneys. It feels productive — even smart. But a landmark federal court decision handed down in February 2026 should give every franchisor pause before typing sensitive business information into a publicly available AI platform.
The Case: United States v. Heppner
In United States v. Heppner, 820 F.Supp.3d 292 (S.D.N.Y. 2026), Judge Jed Rakoff of the Southern District of New York addressed what he described as a question of first impression nationwide: are a user’s communications with a publicly available generative AI platform protected from government inspection by the attorney-client privilege or the work product doctrine?
The defendant, Bradley Heppner, was indicted on charges including securities fraud, wire fraud, and falsifying corporate records. Following his arrest, FBI agents seized approximately 31 documents reflecting his communications with the AI platform Claude. Heppner’s counsel argued those documents should be privileged because: (1) Heppner had inputted information he learned from his attorneys into Claude; (2) he created the documents in order to facilitate discussions with counsel; and (3) he subsequently shared the documents with his lawyers.
Judge Rakoff was unpersuaded. The court ruled that the AI documents were protected by neither the attorney-client privilege nor the work product doctrine — and the reasoning carries direct implications far beyond criminal defense.
Why the Privilege Claims Failed
Attorney-Client Privilege: The privilege requires a confidential communication between a client and an attorney, made for the purpose of obtaining legal advice. The Heppner court found that AI documents fail on multiple grounds:
- Claude is not an attorney. No attorney-client relationship can exist between a user and an AI platform, regardless of how legal the subject matter feels.
- There is no confidentiality. Anthropic, the operator of the Claude AI platform, published its privacy policy — which users agree to — that expressly permits the company to collect user inputs, use them to train its models, and disclose them to third parties, including governmental regulatory authorities. By clicking “agree,” users effectively waive any reasonable expectation of confidentiality.
- The purpose was not to obtain legal advice from counsel. Because Heppner used Claude on his own initiative — without any direction from his attorneys — the court evaluated whether he intended to obtain legal advice from the AI. Claude itself disclaimed that ability, telling the government when asked that it “can’t provide formal legal advice.”
Work Product Doctrine: This doctrine protects materials prepared by or at the direction of counsel in anticipation of litigation. Here again, the court found the AI documents unprotected because Heppner created them entirely on his own initiative, without instruction from his lawyers. The documents reflected his own thinking, not his counsel’s legal strategy — and sharing them with his attorneys afterward did not retroactively create privilege.
The court’s memorable formulation: non-privileged communications are not “alchemically changed into privileged ones upon being shared with counsel.”
What This Means for Franchisors
The Heppner decision arose in a criminal context, but its reasoning applies equally to the civil and transactional settings where franchisors routinely operate. Consider these common scenarios:
Preparing for a franchise disclosure or registration matter. A franchisor’s in-house team might, for example, use an AI platform to draft talking points, research FDD (Franchise Disclosure Document) disclosure obligations, or summarize regulatory requirements before a call with outside franchise counsel. If those AI conversations contain sensitive business information — deal terms, compliance concerns, strategy — they may be discoverable in future litigation or regulatory proceedings.
Evaluating a franchisee dispute. Before engaging counsel, a franchisor’s operations team might run scenarios through an AI chatbot to assess litigation risk, termination options, or settlement strategy. That analysis, however preliminary, likely carries no privilege protection and could be obtained by the franchisee in discovery.
Guiding a franchise transaction, system structuring decisions, or other significant changes. Executives might use AI to model outcomes, draft internal memos, or articulate negotiating positions ahead of attorney consultations. Those outputs typically belong to the AI platform’s operator — and its privacy policy may permit their disclosure.
Practical Guidance for Franchisor Clients
The Heppner decision does not mean franchisors should stop using AI tools. It means they should use them strategically and carefully:
- Involve counsel before using AI for legally sensitive matters. The court left open the possibility that materials prepared at the direction of counsel could receive greater protection. If your attorney asks you to use an AI tool to organize information or draft a summary, that instruction matters legally.
- Treat AI conversations as if they were public. Assume everything you type into a public AI platform could be read by a regulator, opposing counsel, or a jury. Do not input privileged communications, sensitive strategy, or confidential business information that you would not want disclosed.
- Consider enterprise-grade or private AI deployments. Some AI tools offer configurations with stronger data privacy protections that may not share inputs with third parties or train on user data. Your franchise counsel can help evaluate whether such tools offer meaningfully better protection.
- Consult counsel before you prepare, not just after. The most effective attorney-client relationships are proactive. Reach out to your franchise attorney early — before you start researching, drafting, or strategizing — so that your preparations unfold within the protected space that privilege provides.
The Bottom Line
AI is a rapidly developing technology with new implications for how legal matters are approached. United States v. Heppner is a timely reminder that AI tools, however powerful, are not lawyers — and they are not confidential. Franchisors who use publicly available AI platforms to prepare for legal discussions may inadvertently strip away the very protections they are trying to preserve. In franchise law, where regulatory compliance, franchise agreement negotiations, franchisee disputes, and system-wide transactions are never far from potential litigation, that is a risk worth taking seriously.
Have a franchise litigation matter you would like to discuss with an attorney? Call Kevin Shelley at 212-705-0814 or any of the attorneys in our nationally-renowned franchise law practice at Kaufmann Gildin & Robbins LLP.
This blog post is for informational purposes only and does not constitute legal advice. If you have questions about how AI tools may affect your privileged communications, please contact our office.
Attorney advertising. © 2026 KAUFMANN GILDIN & ROBBINS • ALL RIGHTS RESERVED. Disclaimer: The information you obtain at this site is not, nor is it intended to be, legal advice. You should consult an attorney for advice regarding your individual situation. We invite you to contact us and welcome your calls, letters and electronic mail. Contacting us does not create an attorney-client relationship. Please do not send any confidential information to us until such time as an attorney-client relationship has been established.