You are worried, you cannot sleep, and at 2 a.m. you open a chatbot and start typing out everything about the criminal accusation hanging over you. Maybe you asked how to explain a bank transfer, how a jury might see your text messages, or flat out how to beat the charge. It felt private, like talking to yourself. Here is the hard truth that too few people understand: when it comes to AI and attorney-client privilege, that late night conversation may not be protected at all, and prosecutors may be able to read every word.
This is no longer a hypothetical. A federal court has now ruled directly on whether your AI chats are shielded the way a conversation with your lawyer would be, and the answer surprised a lot of people. If you are under investigation anywhere from Los Angeles County to Orange County, understanding this issue now, before you type another prompt, can protect you later. If you have questions about your own situation, do not guess. Call The Law Offices of Christopher Chaney at 818-330-5198 or reach us through the online contact form and let a Los Angeles criminal defense attorney look at the facts before anyone else does.
The Heppner Ruling Explained: Why Did A Federal Court Say AI Chats Are Not Privileged?
In United States v. Heppner, decided in early 2026 in the Southern District of New York, a defendant facing fraud charges turned to a generative AI assistant to help think through his defense. After learning he was the target of a grand jury investigation, he fed information into the AI tool, generated roughly three dozen documents mapping out defense strategy and possible arguments, and later shared some of that material with his attorneys. When the FBI searched his home, agents seized those AI documents. His lawyers argued the material was protected, just like notes prepared for counsel would be.
The court disagreed and granted the government's motion to use the documents. Judge Rakoff found the AI exchanges failed the basic test for attorney-client privilege on more than one ground. The reasoning matters far more than the headline, so here is what the court actually focused on:
- The AI is not your attorney: Privilege protects confidential communications between a client and a lawyer, and the court held that because the AI assistant is not an attorney, that point alone defeated the claim
- There was no real confidentiality: The defendant shared his words with a third-party technology company whose policies allowed the data to be used and even disclosed to outside parties, so the court found no reasonable expectation that the chats would stay private
- Counsel did not direct the AI use: The defendant ran the searches on his own, not at his lawyers' instruction, which meant the material was not created as part of a protected legal workflow
The court left one important door open. It suggested that if an attorney had actually directed the client to use the AI tool, the analysis might have been different, because the tool could then resemble a translator or accountant working under a lawyer's supervision. That distinction is the whole ballgame, and it is exactly why timing and legal guidance matter so much. If you are being investigated in the Los Angeles area and you have been using AI to make sense of your case, talk to us first. Call 818-330-5198 or use the contact form to schedule a consultation.
Similar Post: Is “Pressing Charges” a Real Thing in California? Understanding the Reality of the Criminal Justice System
Your Search History And AI Prompts: Can Prosecutors Use What You Typed After An Indictment?
Think about it the way a seasoned criminal defense lawyer would. If you were indicted and then ran a Google search for how to beat the charge, would that search be discoverable? In most cases, yes. Prosecutors regularly obtain browsing history, search queries, and stored files through search warrants, and courts have allowed that kind of digital evidence for years. Your AI conversations sit in the same category. If a plain search history can be pulled into a case, so can a detailed back and forth with a chatbot where you spelled out your fears, your version of events, and your strategy.
There is a wrinkle many people miss. Several popular browsers and search engines now blend AI directly into ordinary searches, so you may be creating AI-assisted records without realizing it. A few points worth keeping in mind:
- Everyday searches now use AI: Major browsers and search tools increasingly run AI in the background, which means a routine search can generate the kind of AI record at issue in Heppner
- Privacy-focused tools behave differently: Some search engines are built to avoid this kind of tracking and AI integration, though no tool should be treated as a substitute for legal advice
- Deletion is not a fix: Trying to erase searches or chats after you know you are under investigation can create separate and serious problems, including allegations of obstruction
The safest assumption is simple. Treat anything you type into a search bar or a chatbot as something a prosecutor might one day read aloud in a courtroom. After that sinks in, the next section explains what to do instead.
Talking To AI Before You Hire A Lawyer: What Should You Do If You Are Under Investigation In Los Angeles?
The most damaging pattern we see is someone trying to build their own defense through AI in the weeks after they sense trouble, long before they call a lawyer. It feels productive. In reality, it can hand the other side a roadmap. The Heppner court made clear that self-directed AI use, done without a lawyer's involvement, is the version least likely to be protected.
Here are practical next steps if you believe you are being investigated anywhere from downtown Los Angeles to the courthouses along the 405:
- Stop typing about your case: Do not describe the facts, your theories, or your worries to any AI tool, search engine, or app
- Preserve, do not destroy: Leave your devices and accounts as they are, since deleting material after you suspect an investigation can be treated as evidence of intent
- Write nothing you would not want read in court: This includes texts, emails, social posts, and chatbot prompts
- Call a criminal defense attorney early: The sooner counsel is involved, the sooner your communications can be structured to stay within the protection the law actually provides
Getting a criminal defense lawyer involved early does more than calm your nerves. It changes the legal status of the work you do next, which is the practical lesson buried inside the Heppner decision.
Similar Post: What Kind of Proof Is Needed for a Conviction?
California Attorney-Client Privilege And Work Product: What Does The Law Actually Protect?
Heppner is a federal decision from New York, so it is not binding on a California state court. It is, however, one of the first rulings of its kind in the country, and judges here can look to it as persuasive guidance. More importantly, the California rules it echoes point in the same direction, so the lesson travels well to Los Angeles County and the rest of Southern California.
Under California law, the lawyer-client privilege is defined by statute. California Evidence Code section 954 protects confidential communications between a client and a lawyer, and section 952 explains that a communication is confidential when it is made in the course of that relationship and not disclosed to outsiders beyond those helping deliver legal services. Section 953 makes clear that the privilege belongs to the client. The through line in all of these rules is confidentiality and the presence of a lawyer, the two things the Heppner defendant could not show.
A few plain-English takeaways for people facing charges in California:
- A chatbot is not a lawyer: Typing into an AI tool does not create the attorney-client relationship that Evidence Code section 954 requires
- Sharing breaks confidentiality: Giving your words to a third-party platform can waive protection under the confidentiality rules in section 952
- Attorney direction can matter: Work done at the direction of your lawyer stands on very different footing than work you do alone
- No outcome is ever guaranteed: Every case turns on its own facts, and the law in this area is still developing
None of this means technology has no place in a defense. It means the safeguards depend on doing things in the right order, with counsel guiding the process rather than discovering it after the fact.
Facing Criminal Charges?
We Can Help.
Speak With a Top-Rated Criminal Defense Attorney Today
818-330-5198Facing Charges In Los Angeles: How Can A Criminal Defense Attorney Help Protect Your Rights?
When you bring your situation to a lawyer early, several things change at once. Your conversations with counsel are covered by the privilege the Heppner defendant lacked. Any research or analysis can be organized under attorney direction, which gives it the best chance of protection. And you get honest guidance on what to say, what to preserve, and what to stop doing right now.
Christopher Chaney has spent years defending people across Los Angeles, from Encino and the San Fernando Valley to Century City and the courthouses off Ventura Boulevard. The firm handles serious matters with a calm, focused, and determined approach, and we understand how quickly digital evidence, including AI records, can shape a case. Technology moves fast, and the law is racing to keep up, which is exactly why you want experienced counsel who is paying attention to rulings like Heppner as they land.
If you are worried about something you already typed, or you simply want to make sure your next move is the right one, reach out before the situation gets harder to manage. Call The Law Offices of Christopher Chaney at 818-330-5198or use the secure online contact form to request a consultation.
Frequently Asked Questions About AI And Attorney-Client Privilege
Are my ChatGPT or AI chats protected by attorney-client privilege?
Generally no. As the Heppner ruling showed, an AI tool is not an attorney, and sharing your words with a third-party platform usually means there is no reasonable expectation of confidentiality. That combination places most AI chats outside the privilege.
Can prosecutors get my AI conversations and search history in a criminal case?
Often yes. Prosecutors can seek digital records, including browsing history, search queries, and AI chats, through search warrants and other legal processes, much the way they obtain other stored data.
Is it safer to use a privacy-focused search engine?
Some search tools avoid the AI integration built into mainstream browsers, but no tool replaces legal advice. The safest step if you are under investigation in Los Angeles is to stop discussing your case online and speak with a criminal defense attorney.
What should I do if I already talked to an AI tool about my case?
Do not delete anything, since erasing material after you suspect an investigation can create new problems. Contact a Los Angeles criminal defense attorney right away so your situation can be assessed and handled properly.
Protect Yourself Before You Type Another Word
The Heppner decision is a warning for anyone who has leaned on AI while facing a criminal accusation. What felt like a private brainstorm can become evidence, and the protection you assumed you had may not exist. The good news is that the right guidance, early enough, still makes a real difference in how your defense takes shape.
If you are under investigation or facing charges anywhere in the Los Angeles area, do not navigate this alone and do not rely on a chatbot to defend you. Call The Law Offices of Christopher Chaney at 818-330-5198or reach out through the online contact form to speak with a Los Angeles criminal defense attorney who can help you protect your rights.
Disclaimer: This blog is intended for informational purposes only and does not establish an attorney-client relationship. It should not be considered as legal advice. For personalized legal assistance, please consult our team directly.
