AI Communications Are Not Privileged: What United States v. Heppner Means for Corporate Clients.

The article focuses on a February 17, 2026, federal court decision in United States v. Heppner, where Judge Jed S. Rakoff ruled that written communications between a criminal defendant and a generative AI platform, specifically Anthropic's Claude AI, are not protected by attorney-client privilege or the work-product doctrine. The court held that AI platforms are considered third parties without fiduciary duties, that users lack a reasonable expectation of confidentiality due to AI privacy policies, and that sharing information with AI waives any privilege. Additionally, the work-product doctrine does not apply because the AI-generated materials were not prepared by or at the direction of counsel. This ruling signals significant implications for corporate clients and attorneys using AI in legal contexts, emphasizing that AI-generated documents may be discoverable and that careful management of AI use is necessary to protect confidentiality and privilege. [Extracted from the article]