The Expert RecordThe standard for expert witnesses using AI

AI in the Courtroom · Judges, courts and AI

Senate Judiciary inquiry into judges' AI use

Senate Judiciary Chairman Chuck Grassley asked Judges Wingate and Neals whether AI was used in drafting error-filled orders. Both acknowledged staff use, and the Administrative Office of the U.S. Courts described its interim AI guidance.

Primary document readJudges acknowledged staff AI use; AO described interim guidance; chairman called for permanent policy October 23, 2025

By · Published · Updated · How this site is researched

What happened

By late summer 2025, 2 federal district judges had withdrawn rulings containing errors of the kind generative AI produces. Judge Julien Xavier Neals of the District of New Jersey withdrew an opinion in the CorMedix securities case. Judge Henry T. Wingate of the Southern District of Mississippi replaced a temporary restraining order in a challenge to the state's DEI law. Neither judge had explained publicly how the errors arose.

On October 6, 2025, Senate Judiciary Chairman Chuck Grassley wrote to both judges. He asked whether they, their law clerks or court staff had used generative AI in preparing the decisions, and whether any non-public case information had been entered into an AI tool. He also asked them to restore the original orders to the docket. His release stated: "No less than the attorneys who appear before them, judges must be held to the highest standards of integrity, candor, and factual accuracy."

What the judges and the judiciary said

Judge Neals responded on October 20, 2025, stating that a law school intern had used ChatGPT without authorization. Judge Wingate responded on October 21, 2025, stating that a law clerk had used Perplexity to synthesize docket information. Both described the docketed rulings as early drafts that had skipped the chambers' normal review, and both described new review procedures.

Robert J. Conrad Jr., Director of the Administrative Office of the U.S. Courts, also wrote on October 21, 2025. He described an AI Task Force of judges, court executives, and technology and chambers staff, and interim guidance distributed to the courts on July 31, 2025. According to his letter, the guidance cautions against delegating core judicial functions to AI, including decision-making, and advises users to "review and independently verify all AI-generated content or output." It states that users are "accountable for all work performed with the assistance of AI" and asks judges to consider whether AI use should be disclosed. The letter described the guidance as interim, to be followed by more permanent policy.

Where it stands

On October 23, 2025, Senator Grassley released the responses. He stated that "the judicial branch needs to develop more decisive, meaningful and permanent AI policies and guidelines." No discipline of either judge has been reported. In March 2026, bills titled the Research and Oversight of AI in Courts Act of 2026 were introduced in both chambers, addressing AI speech-to-text technology in the federal courts. A March 2026 survey report published by the New York City Bar Association found that more than 60% of responding federal judges had used at least one AI tool.

On September 17, 2026, the Judicial Conference of the United States received an update from its AI Task Force, chaired by Judge Sidney R. Thomas of the Ninth Circuit and appointed in 2025. According to the judiciary's account of the session, the task force "identified more than 60 distinct issues, prioritized them for review, and formed seven subject-matter subgroups to guide further study." The Director reported that "courts have been cautioned not to delegate core judicial functions to AI, including decision-making or case adjudication," and that "all Judiciary users have been reminded that they are accountable for all work performed with the assistance of AI." The judiciary said it will roll out additional guidance as issues are resolved, and gave no completion date.

So the position as of September 17, 2026 is unchanged in substance: interim guidance distributed July 31, 2025, never published in full, and a permanent judiciary-wide policy still under study more than a year later.

The obligation it engaged

The interim guidance, as the AO letter describes it, tracks 3 clauses of the Expert Record Standard. It addresses non-delegation of decisions (Clause 2), independent verification of output (Clause 3) and consideration of disclosure (Clause 8). The chairman's question about non-public case information goes to Clause 6, custody and confidentiality: what may be submitted to a system at all.

The inquiry also shows what happens when there is no contemporaneous record (Clause 7). Each judge had to reconstruct, months later and under congressional inquiry, who used which system and for what. An expert who keeps a log as the work proceeds can answer the same questions from a document.

Sources

What could not be confirmed

The interim guidance itself was not published in full in the sources opened; its content is taken from the AO letter. The ABA Washington Letter attributes the March 2026 judicial survey to Northwestern University, while the New York City Bar Association page presents it as its task force's report; the entry does not attribute authorship beyond the Bar's publication. Whether a permanent policy was adopted after these sources is not confirmed.

About this entry

This entry was checked against the primary document listed in the sources. The entry for Senate Judiciary inquiry into judges' AI use records conduct and what the tribunal did. It does not characterize a holding and it is not legal advice. Last checked September 17, 2026. More matters like this one are collected under AI errors in judicial orders. If something here is out of date, write to [email protected].

Top