Courts shield AI prompts and outputs from discovery as work product

In June 2026, courts in Texas and New York established that generative artificial intelligence prompts and outputs used for litigation preparation are protected work product. In Assini v. Hayward, a New York court shielded a pro se litigant's AI queries, while in Tate Group Automotive, a Texas court protected a party principal's ChatGPT sessions. However, these protections do not extend to testifying experts' prompts, which remain discoverable methodology, nor do they shield underlying documents uploaded to the tools.

AI work-product protection doctrine

• ▪In June 2026, courts in Texas and New York issued the first reasoned decisions establishing that generative artificial intelligence prompts and outputs used for litigation preparation are protected under work-product doctrines.
• ▪The emerging legal doctrine protects confidential litigation strategy developed via artificial intelligence, but does not shield the underlying documents uploaded into the artificial intelligence tools.

Assini v. Hayward ruling

• ▪On June 4, 2026, the Supreme Court of New York, Nassau County, in Assini v. Hayward, quashed non-party subpoenas seeking a self-represented defendant's artificial intelligence prompts, uploads, and outputs.
• ▪In Assini v. Hayward, the court ruled that a self-represented litigant can claim work-product protection for artificial intelligence-assisted case preparation and that using a commercial artificial intelligence tool does not waive confidentiality.

Tate Group Automotive ruling

• ▪In Tate Group Automotive, LLC v. Legacy Automotive Capital, LLC, Judge Grant Dorfman ordered the plaintiff to disclose all discovery materials and work products that had been shared with ChatGPT.
• ▪On June 3, 2026, the Texas Business Court held in Tate Group Automotive, LLC v. Legacy Automotive Capital, LLC that ChatGPT conversations conducted by a party principal were protected attorney work product.

Expert discovery AI exceptions

• ▪The court in Conservation Law Foundation, Inc. v. Shell Oil Co. reasoned that the expert's artificial intelligence prompts were discoverable methodology under Rule 26 rather than protected notes or drafts.
• ▪In Conservation Law Foundation, Inc. v. Shell Oil Co., a Connecticut federal magistrate judge ordered a party to produce the generative artificial intelligence prompts used by its testifying expert.

AI litigation governance practices

• ▪Litigants are advised to segregate litigation artificial intelligence use into dedicated accounts and threads to strengthen the showing that the materials were created because of litigation.
• ▪Litigants are advised to utilize enterprise artificial intelligence deployments with no-training commitments and confidentiality terms to improve the waiver analysis over consumer-grade tools.