Inspiring dismay in litigators nationwide, in the now-infamous case of Mata v. Avianca, two New York lawyers submitted “non-existent judicial opinions with fake quotes and citations created by the artificial intelligence tool ChatGPT” to the federal district court and continued to “to stand by the fake opinions after judicial orders called their existence into question.”1
The public release of ChatGPT in November 2022 brought massive attention to “generative” artificial intelligence tools that can generate texts in response to prompts.2 Only three months later, the offending submission in Mata was filed, and one of the sanctioned attorneys explained that he “had not previously used ChatGPT” and only “became aware of it through press reports and conversations with family members.”3
The spectacular and well-publicized failure of the attorney’s reliance on ChatGPT in Mata raised awareness of the potential uses and misuses of artificial intelligence tools in legal drafting and stimulated discussion by legal practitioners, theorists, and judges about the impact on the law of the seemingly inevitable further development of generative technology.4