New Guidance Highlights AI Risks in Expert Witness Testimony
New legal guidance highlights discovery risks of AI-generated expert testimony in litigation.
Why it matters: AI use in litigation is growing, creating new evidentiary and ethical challenges for legal professionals. Understanding these risks is vital to managing AI-driven expert analysis in court.
- In Morgan v. V2X, Inc. (2026), a court allowed AI use with confidentiality protections under work product rules.
- Courts increasingly require disclosure of AI prompts and outputs used by expert witnesses, treating them as factual bases for testimony.
- The Academy of Experts released February 2026 guidance urging transparency and understanding of AI tools by expert witnesses.
- Experts' AI prompts create a time-stamped record that may reveal intended outcomes before reports are finalized, per legal analysis.
Recent court decisions and expert guidance are shaping how AI, particularly large language models (LLMs), are integrated into expert witness testimony in litigation. These developments underscore emerging discovery risks and procedural challenges for litigators.
In Morgan v. V2X, Inc., decided on March 30, 2026, Magistrate Judge Maritza Dominguez Braswell approved protective order language that permits AI use during litigation, provided confidentiality safeguards are in place. The court recognized a pro se litigant's use of AI in litigation preparation as work product under Federal Rule of Civil Procedure 26(b)(3) but clarified that this protection does not extend to revealing the identity of the AI tool used.
Conversely, in Conservation Law Foundation, Inc. v. Shell Oil Co., Magistrate Judge Thomas O. Farrish ordered production of AI prompts used by an expert witness to prepare her report. This ruling highlights a growing judicial view that AI-generated prompts and outputs comprise part of the factual basis for expert testimony and are thus subject to discovery.
The Academy of Experts published guidance in February 2026 stressing the importance of AI transparency for expert witnesses. They encourage experts to thoroughly understand AI tools and openly disclose their use to uphold credibility and withstand legal scrutiny.
Legal expert David A. Shargel notes that each AI prompt creates a "time-stamped record" that reveals what the expert sought and sometimes the desired outcome before finalizing a report. This new form of recorded input may significantly impact discovery and evidentiary evaluation processes.
These cases and guidance collectively signal a shift toward greater transparency and cautious handling of AI in expert testimony. Legal professionals should prepare for increased discovery demands related to AI-generated materials and carefully assess ethical considerations surrounding AI tools in litigation.
By the numbers:
- March 30, 2026 — Date of Morgan v. V2X, Inc. court decision
- February 2026 — Publication of Academy of Experts' guidance on AI use in expert testimony