D.C. Appeals Court Strikes Brief Containing AI-Generated Fake Cases

The District of Columbia Court of Appeals has struck a legal brief submitted by attorneys representing a Deutsche Bank subsidiary after discovering the document contained fake case citations. The ruling serves as a stark warning to the legal profession regarding the unverified use of generative artificial intelligence in court filings.

The brief, filed on behalf of Deutsche Bank National Trust Company, included four nonexistent citations that had been generated by an AI tool. Upon discovering the fabrications, the court took the decisive step of striking the brief from the record. In its order, the court explicitly described the incident as a "cautionary tale" regarding the misuse of AI technology in the pursuit of legal research.

The Rise of AI Hallucinations

This incident is part of a broader, growing trend of "AI hallucinations" within the legal industry. Generative AI tools, while capable of synthesizing vast amounts of data, are known to occasionally invent facts, quotes, or—in this case—entire legal precedents that appear authentic but have no basis in reality. When lawyers rely on these tools to conduct research without rigorous manual verification, they risk submitting fraudulent material to the court, often inadvertently.

Risks to Legal Practice

The consequences of such errors extend beyond the immediate loss of a legal argument. The striking of a brief can jeopardize a client's position in a case and expose attorneys to severe professional repercussions. Relying on generative AI without a strict verification process can lead to judicial sanctions, formal reprimands, and in extreme cases, disbarment. This case underscores that the duty of competence and the responsibility for the accuracy of a filing remain solely with the signing attorney, regardless of the tools used to produce the draft.

The Path Forward

As AI tools become more integrated into legal workflows, courts are increasingly vigilant about the provenance of citations. Legal professionals are now being urged to implement mandatory "human-in-the-loop" verification for every case cited in a filing. While the court's order in the Deutsche Bank matter highlights the dangers of automation, it also signals a period of transition where the judiciary will likely impose stricter disclosure requirements for AI-assisted work to prevent further instances of fabricated law.