Federal and state courts in New Mexico have increasingly encountered legal filings containing false or misleading information generated by artificial intelligence (AI), leading to sanctions and judicial reprimands. Since 2023, at least seven lawsuits in the state have featured what judges describe as "AI hallucinations," a phenomenon where AI tools produce fabricated or incorrect content.

One notable case involved a 62-year-old man from Albuquerque who filed a federal lawsuit without legal representation, claiming lifelong disability discrimination and retaliation by a former employer. Among his damages request was an extraordinary monetary sanction totaling $355.69 quintillion—an amount Senior U.S. District Judge Judith Herrera deemed “ludicrous.” Ultimately, the plaintiff was ordered to pay $8,640 in sanctions after the court found elements of his filings were tainted by AI-generated inaccuracies.

Federal judges have highlighted the growing challenge of ensuring the accuracy of pleadings and motions as more litigants and some attorneys rely on generative AI tools, such as ChatGPT, to draft legal documents. In one instance, U.S. Magistrate Judge Damijan Martinez of Las Cruces identified a legal brief that contained references to six entirely fictitious cases, which were likely the product of AI-generated "hallucinations." These false citations were submitted by an attorney based outside New Mexico and were not reviewed by a local lawyer before filing.

In his ruling, Martinez described AI hallucinations as resulting from the predictive nature of AI models. These systems analyze vast datasets to generate text but can "hallucinate" by inventing plausible yet false information, including case law or legal authorities that do not exist. This issue raises concerns about the reliability of AI-assisted legal drafting and the risk of misleading courts.

The emergence of AI-generated errors has prompted calls within the legal community for greater scrutiny of documents produced with AI assistance, particularly in pro se cases where litigants lack formal legal training. Judges are increasingly issuing warnings and imposing sanctions to deter reliance on unchecked AI content and to uphold the integrity of judicial proceedings.

While AI technologies continue to transform several aspects of legal practice, the recent cases in New Mexico highlight the necessity of human oversight to prevent the admission of fabricated or erroneous information in court filings. The judiciary’s response reflects an effort to balance innovation with the fundamental requirement of accurate, verifiable legal arguments.