The number of cases discussing the use of generative AI produced hallucinated content – typically fake citations, but also other types of AI-generated arguments keeps growing. According to one database, as of today, that number is 1571 – but it is likely to be higher by the time you read this.
And, as you probably have also noticed, as the number of cases goes up, the patience level of judges goes down. I have noticed more judges imposing sanctions (rather than warnings) and that those sanctions are becoming harsher.
Inevitably, this will lead to several questions: under what circumstances should judges conclude that the mistakes they encounter are, in fact, the result of the use of AI, and what is the appropriate response or sanction?
A couple of weeks ago, Above the Law published a short article exploring this issue. In it, the authors argues that just because a judge thinks a lawyer made a bad argument it does not mean that the judge can mean legitimately conclude the lawyer relied on AI; and that, even if it did, maybe the conduct should not be sanctionable. He argues that judges reactions to bad lawyering may be turning into a dangerous slippery slope.
The article is called AI Slop Or Novel Arguments? A Texas Judge Goes Off The Rails and you can read it here.
No comments:
Post a Comment