A Delaware court has sanctioned an attorney $2,500 for submitting a legal brief that included fabricated quotes and mischaracterized legal authorities generated by a generative artificial intelligence (GenAI) tool. This ruling highlights the growing concerns about the use of AI in legal practice and the responsibility attorneys have to ensure the accuracy of their submissions.

The case, Kevin Leiske et al. v. Robert Gregory Kidd et al., was heard in the Court of Chancery of Delaware under docket number 2025-0426-CDW. The ruling was issued by Vice Chancellor Lori W. Will on November 13, 2026. The court's decision comes amid increasing reliance on AI tools in various professions, including law, where the accuracy of information is paramount.

The dispute arose when the plaintiffs submitted an answering brief opposing the defendants' exceptions to a Magistrate in Chancery's final report on advancement. Shortly after, the defendants identified significant errors in the plaintiffs' brief, including fabricated quotations and inaccurate citations. The plaintiffs' lead counsel acknowledged using a GenAI tool to assist in drafting the brief, but failed to verify the output before filing.

In the court's opinion, it was noted that the use of GenAI in legal practice is not inherently problematic, but attorneys must ensure the accuracy of all citations and legal authorities presented to the court. The court stated, "Counsel failed to verify the accuracy of the answering brief despite his law firm’s comprehensive GenAI policy and mandatory ethics training." This failure to verify led to the imposition of sanctions against the individual attorney, while the law firm was not sanctioned due to its proactive measures to prevent such misuse.

The court emphasized that an attorney's signature on a court filing is a certification to the tribunal that the legal contentions made are warranted by existing law. The ruling outlined that presenting unverified GenAI output violates Rule 11(b)(2) of the Court of Chancery, which requires attorneys to conduct a reasonable inquiry into the validity of their claims.

In this case, the plaintiffs' lead counsel, who also served as the Chair of the law firm's AI Subcommittee, had used an unapproved GenAI tool to help draft the brief. He delegated the task of cite-checking to a junior attorney and a paralegal, which ultimately led to the submission of inaccurate information. The court noted that while reliance on junior staff for cite-checking is common, the responsibility for ensuring accuracy remains with the signing attorney.

Vice Chancellor Will pointed out that the duty of candor to the court and the obligation to conduct a reasonable inquiry are personal and cannot be delegated. The attorney's failure to verify the information before submission resulted in the court's decision to impose a $2,500 sanction, which he must pay personally within 14 days.

The ruling serves as a cautionary tale for attorneys who utilize AI tools in their practice. The court highlighted that the use of GenAI can offer significant benefits but also carries risks if not managed properly. It stated, "The use of GenAI in legal practice is neither forbidden nor inherently problematic," but emphasized the need for careful oversight and verification.

Moving forward, this ruling may have implications for how attorneys approach the use of AI tools in their work. It underscores the necessity for legal professionals to maintain rigorous standards of accuracy and accountability, especially as technology continues to evolve in the legal field. The court's decision may prompt law firms to review and strengthen their policies regarding the use of AI in legal drafting and research.

As for the future of this case, it is unclear whether the attorney intends to appeal the ruling or if there are related cases pending. The court filing did not provide additional details on any potential appeals or further legal actions.