A word of caution: Do not use public AI tools to research, draft, or “test” your defense if you are under investigation or facing charges in New Mexico.
A groundbreaking 2026 federal court decision confirmed that AI prompts and outputs are not protected by attorney-client privilege. Law enforcement can seize those communications, read your defense strategy, and use your own words against you in court even If you type the facts of your case into an AI chatbot.
In This Artcle
Criminal defense lawyer, Marc A. Grano, shares the primary issues surrounding “AI Lawyers” in New Mexico and how they can seem like the most affordable choice in a moment of panic but will cost you down the procedural line. Grano Law Offices, P.C. wrote this post to ensure that every New Mexico citizen and family understands how this long-standing legal within the field of law.
Table of Contents
The Dangerous Illusion of the “AI Lawyer”
Anxiety can be more than overwhelming when facing a criminal investigation. It is natural to want immediate answers about:
- Potential charges,
- Specific laws, and;
- Your legal exposure.
Increasingly, individuals are turning to public artificial intelligence platforms like ChatGPT, Claude, and Gemini to outline their defense strategies or anticipate how prosecutors might attack their case when they encounter their tactics.
This mistake is.catastrophic.
Public AI is not your lawyer, it is not confidential, and using it can hand prosecutors your exact words, timelines, and strategy on a silver platter.
United States v. Heppner: The Government Can Read Your Prompts
If you believe your AI chats are private, a recent federal court decision serves as a terrifying reality check. In February 2026, a federal judge in the Southern District of New York issued a landmark ruling in United States v. Heppner.
After learning he was the target of a federal fraud investigation, the defendant used a public AI platform (Claude) to create written “reports.” He typed in prompts outlining his potential defense strategy, his view of the facts, and his anticipated responses to the government’s charges.
When the FBI eventually arrested him and executed a search warrant at his home, they seized his electronic devices. Among the evidence collected were 31 documents memorializing his exact conversations with the AI.
The defense team fought to keep these documents out of the trial, arguing they were protected by attorney-client privilege because the defendant had used the AI to prepare to speak with his lawyers. The federal judge firmly rejected this argument, and AI documents were handed over to the government.
Why AI Fails the Test Under Rule 11-503 NMRA
In New Mexico, attorney-client privilege is strictly governed by Rule 11-503 NMRA. For a communication to be legally protected from law enforcement, it must be a confidential communication made strictly for the purpose of facilitating or providing legal services.
As the judge in Heppner pointed out, interacting with a public AI chatbot fails the legal test for privilege for several critical reasons:
- AI is Not an Attorney: Attorney-client privilege requires a licensed professional who owes you a fiduciary duty. A software program cannot hold privilege.
- You Are Acting Unrepresented: Unless your retained defense attorney explicitly directs and supervises your use of a secure, closed-network AI tool, your independent use of consumer AI is treated legally as you talking to a stranger
The Illusion of Privacy and Confidentiality
perhaps the most dangerous aspect of public AI platforms is the fine print. When you use tools like ChatGPT, Claude, or Gemini, you consent to their Terms of Service (ToS). These policies explicitly state that the companies collect your data, use your inputs to train their models, and reserve the right to disclose your data to third parties—including government regulatory authorities and law enforcement.
The moment you hit “enter” on a prompt containing details about an alleged crime, you have legally waived your right to confidentiality.
Safer Alternative: Hire a Criminal Defense Lawyer
When building an aggressive criminal defense strategy, information control is paramount. If you must use technology during your case, strictly adhere to the following protocols, and only with your lawyer’s explicit approval:
- Use AI for General Definitions Only: You may use AI to ask purely general, educational questions (e.g., “What is a preliminary examination in New Mexico?”). Never enter facts, dates, names, or specifics related to your actual case.
- Rely on Firm-Vetted Tools: Ask your lawyer if they utilize enterprise-grade, secure legal AI tools that do not train on user data. Any AI-assisted document review must happen at counsel’s direction and within agreed protocols.
- Keep Communications Internal: Use secure, lawyer-approved portals for sharing documents and timelines. Avoid copying anything into public chatbots to check spelling or grammar.
The Rule of Thumb: Keep AI Out of Your Case
The key takeaway is simple: Do not enter case facts into public AI. Do tell your lawyer immediately if you previously used AI to discuss the case. And always assume AI chats can be seen, seized, and used by the state.
If you are facing complex state felony charges or are under investigation, do not trust a chatbot with your freedom.
Start with Our Free Case Strategy Session
Contact Grano Law Offices, P.C. today at (505) 426-8711 or use the live chat on this page to schedule a confidential Free Case Strategy Session with a real, licensed New Mexico defense attorney.
