Landmark AI Rulings Will Have Effect on All Litigation

Why It Matters

The rulings clarify how privilege and discovery rules apply to AI‑generated content, guiding law firms and litigants in protecting sensitive strategy while avoiding inadvertent disclosures.

Key Takeaways

• Supervised AI use retains work‑product protection.
• Unsup­ervised AI communications lack privilege.
• Platform privacy policies dictate confidentiality expectations.
• Document attorney direction to safeguard AI‑generated work.
• Public AI tools risk seizure in investigations.

Pulse Analysis

Generative AI has moved from experimental novelty to a routine component of legal research, drafting, and case strategy. As firms integrate large‑language models into their workflows, courts are forced to reconcile existing evidentiary doctrines with these new tools. The recent decisions from Michigan and New York signal that the judiciary will not create novel privileges for AI output; instead, it will apply established standards—work‑product protection and attorney‑client privilege—based on how the technology is used and the surrounding facts.

In Warner v. Gilbarco, the court emphasized that AI programs are merely tools, not parties, and that the work‑product doctrine protects an attorney’s mental impressions when the AI is employed under direct supervision. Because the request sought Warner’s internal thought process rather than tangible documents, the court deemed it a fishing expedition and denied discovery. This outcome reinforces that maintaining attorney control and documenting AI‑assisted steps can preserve the confidentiality of strategic materials, even when the software itself processes the information.

The Heppner case illustrates the opposite scenario. The defendant used a public Anthropic model without counsel’s direction, and the platform’s privacy policy allowed data retention and third‑party disclosure. The court ruled that no reasonable expectation of confidentiality existed, so neither privilege nor work‑product protection applied. Practitioners should therefore adopt written AI‑use protocols, select enterprise‑grade tools with contractual data safeguards, and raise AI‑related issues early in discovery conferences. By aligning technology choices with traditional privilege requirements, lawyers can harness AI’s efficiency without exposing critical litigation assets to adverse parties.