Proposed Federal Rule of Evidence 707: What Business and Their Lawyers Need to Know
May 27, 2026
Courts have adapted evidentiary rules to new technology for decades. With artificial intelligence now producing litigation-ready analysis at scale, Proposed Federal Rule of Evidence 707 signals that formal rulemaking is catching up.
What Is Proposed Rule 707?
Proposed Rule 707 would require certain machine-generated evidence, including AI output, to satisfy the same admissibility standards as expert testimony under Rule 702 when offered without a sponsoring expert.
In practical terms, the rule would prevent parties from bypassing expert-witness scrutiny when presenting AI output directly. If an AI system generates conclusions that would require a human to testify as an expert, the proponent must show the output is based on sufficient facts or data, produced using reliable principles and methods, and reflects a reliable application of those methods to the case. This effectively applies the familiar Daubert framework to machines when they function like experts.
The Advisory Committee on Evidence Rules voted 8 to 1 in May 2025 to publish Rule 707 for public comment. The comment period closed February 16, 2026, and a final report is expected in June 2026, after which the proposal would proceed through Supreme Court transmission and potential congressional review.
Why It Matters
Rule 707 responds to a growing pattern of failures in which lawyers treated AI output as reliable without verification.
The catalyst was Mata v. Avianca, Inc., 678 F. Supp. 3d 443 (S.D.N.Y. 2023), where attorneys submitted a brief citing judicial decisions that did not exist, fabricated by ChatGPT, complete with plausible captions and quotations. Even after opposing counsel challenged the citations, the lawyers relied on the AI’s assurances that the cases were real. They were not. The court imposed sanctions and found subjective bad faith.
Mata was not an outlier. By mid-2025, courts around the country were confronting a steady stream of filings containing hallucinated AI citations. Judicial patience with counsel blaming the software rather than themselves quickly wore thin.
This shift was crystallized in Johnson v. Dunn, 792 F.Supp.3d 1241(N.D. Ala. 2025). Despite firmwide AI policies, lawyers at a major firm filed motions containing AI-fabricated citations. The court declined to sanction the firm itself but imposed severe sanctions on individual attorneys, including public reprimands, disqualification, and referrals to the state bar disciplinary board, emphasizing that accountability follows the attorney’s signature, not the tool.
The lesson from these decisions is not that AI is unusable, but that unverified AI output is professionally hazardous.
What Rule 707 Does and Does Not Do
Rule 707 addresses a narrow but important evidentiary gap. Under existing rules, machine-generated conclusions can sometimes be offered as standalone evidence even when they perform the same analytical function as expert testimony. Rule 707 closes that loophole by tying admissibility to function rather than form.
At the same time, the proposal is deliberately limited. It applies only to evidence the proponent acknowledges is machine-generated. It does not solve authentication problems such as deepfakes or undisclosed AI use. And it leaves undefined the boundary between complex AI analysis and “simple scientific instruments,” ensuring that the scope of the rule will likely be subject to early litigation.
Practical Implications
For litigators, Rule 707 will likely mean additional pretrial admissibility disputes, including hearings focused on training data, validation, and methodological reliability.
For business clients and transactional counsel, the takeaway is simpler: AI output does not shift professional responsibility. Whether AI is used for litigation strategy, damages modeling, contract analysis, or regulatory submissions, lawyers remain responsible for verifying accuracy and reliability.
Rule 707 is not anti-AI. It does not prohibit machine-generated evidence or discourage innovation. On the contrary, it legitimizes the potential use of AI in the courtroom by enforcing a principle courts already apply to humans: if an opinion influences fact-finding, it must be reliable enough to withstand scrutiny.
Click here to view a downloadable PDF of the legal update.
This Legal Update is intended to keep readers current on developments in the law. It is not intended to be legal advice. If you have any questions, please contact Derek Luke at 412.566.6967 or dluke@eckertseamans.com, Shane Valenzi at 412.566.6186 or svalenzi@eckertseamans.com, or any other attorney at Eckert Seamans with whom you have been working.
Share This Post
Authors
