Question · Voice privacy and security
Are conversations with an AI chatbot privileged?
Probably not, at least for a consumer chatbot a non-lawyer uses on their own. In February 2026 a federal court in New York held that a defendant's conversations with Claude were not protected by attorney-client privilege or work product. The court did not decide how a tool used at a lawyer's direction would be treated.
Updated October 17, 20263 min read5 cited facts, each traced to its source
Not as a general rule. Attorney-client privilege protects confidential communications between a lawyer and a client made to get legal advice 1. A chatbot is not a lawyer, and a conversation with one is not automatically confidential. The one federal ruling we found on this point treated a person's own conversations with a consumer AI tool as unprotected. That ruling is recent and tied to its facts, so it is a warning, not a settled rule for every case.
A federal court held that consumer chats were not privileged
On February 13, 2026, Judge Jed S. Rakoff of the Southern District of New York ruled that a non-attorney's communications with a generative AI platform were not protected by attorney-client privilege or work product 2. The case is United States v. Heppner. Commentators differ on some details, so check the docket before you cite it.
The court's basic reasoning was that two non-attorneys discussing legal issues are not in a privileged relationship, however accurate the exchange is. The court also looked at the terms of the tool, discussed below.
The defendant used the tool on his own
The facts limit the ruling. The defendant used Claude, a publicly available AI assistant, to analyze his legal situation and develop defense arguments, and he did so "on his own initiative, without direction from his attorneys" 3. Because the defendant acted alone, the court did not decide what happens when a lawyer directs an AI tool as part of representing a client. That question is open.
The privacy policy weighed against confidentiality
Orrick, a law firm that analyzed the ruling, says part of the court's reasoning turned on Anthropic's consumer privacy policy. Under that policy the company uses user data for training and reserves the right to disclose it to third parties, including government authorities 4. A user who accepts those terms has a hard time claiming a reasonable expectation of confidentiality. Orrick also notes that enterprise AI platforms with negotiated confidentiality terms may present a different picture, though no court has tested that yet.
Chat history is a record that lasts
A chatbot conversation is also a stored record. OpenAI says audio from ChatGPT dictation is kept for as long as the chat stays in your chat history, and deleting the chat removes it 5. The same logic applies to typed chats. If the tool keeps the conversation, the conversation can be requested or reviewed, and the privacy argument gets weaker.
It depends on five things
- Who the user is. Privilege needs a lawyer in the relationship. A client using a chatbot alone has no lawyer in the conversation.
- Whether a lawyer directed the use. Heppner stressed that the defendant acted without his attorneys' direction. A lawyer who directs an AI tool to help with a client's matter stands on different ground, but the courts have not ruled on that yet.
- Which tool and which terms. A consumer product with broad rights to use and disclose data is much weaker than an enterprise tool with negotiated confidentiality.
- Where the record lives. A tool inside the firm's controlled environment is easier to defend than a consumer account.
- Whether the output goes to counsel later. Commentary on Heppner reads the court as rejecting the argument that sending AI output to a lawyer afterward makes it privileged. Do not rely on that step to protect material, and check the opinion itself.
What to do if you use a chatbot on a legal matter
- Do not use a consumer chatbot for case strategy on a client matter unless your firm has reviewed its terms and approved it.
- Read the tool's data-use and disclosure terms. Check whether it trains on your inputs, how long it keeps them and whether it can disclose them to third parties.
- Use enterprise tools under written confidentiality terms where you can, and keep the lawyer in control of the work.
- Treat chat logs as discoverable records. Assume that anything you type could be produced in litigation.
For the broader question of how dictation and AI tools affect privilege, see our guide on dictation for lawyers, privilege and confidentiality. Related questions cover whether cloud dictation waives attorney-client privilege, what ABA Formal Opinion 512 says about AI tools and whether AI models can leak what you dictated. For the definition of the protection, see attorney-client privilege.
This is general information about one ruling, not legal advice. Check the docket and your jurisdiction's rules before you rely on it for a matter.
Evidence
Every number on this page, with where it comes from. Open a fact to see the exact words in its source.
- 1.
- 2.
- 3.
- 4.
- 5.
Go deeper
Related questions
- Can lawyers use AI notetakers on client calls?
- Does using cloud dictation waive attorney-client privilege?
- What does ABA Formal Opinion 512 say about AI tools?
- Are open-source offline dictation apps like Handy and Murmur safe to use?
- Can a third-party keyboard see what you type?
- Can AI models leak what you dictated?
- Can apps read your clipboard?
- Can government employees use AI dictation?
Referenced by
- Speech to Text Privacy: Where Your Voice GoesTopic
- Dictation for Lawyers: Privilege and ConfidentialityGuide
- Can lawyers use AI notetakers on client calls?Question
- Does using cloud dictation waive attorney-client privilege?Question
- What does ABA Formal Opinion 512 say about AI tools?Question
- Attorney-client privilegeGlossary