AI models frequently ‘hallucinate’ on legal queries, study finds
Generative artificial intelligence (AI) models frequently produce false legal information, with so-called “hallucinations” occurring between 69 percent to 88 percent of the time, according to a recent study. Large language …