A study of 100 recent U.S. cases involving AI-related inaccuracies in legal filings found that while the seriousness of the original error remains central—with nearly every high-materiality incident producing a serious consequence—a lawyer's post-discovery conduct also matters. Serious consequences were substantially more common when lawyers offered misleading explanations, shifted blame, repeated the misconduct, or introduced additional inaccuracies after receiving notice. Prompt admission was associated with a markedly lower rate of serious consequences. The study, which began with Damien Charlotin’s publicly available AI Hallucination Cases Database and independently reviewed 100 of the most recent resolved U.S. matters involving lawyers, examined what courts did after lawyers learned that AI-related inaccuracies had entered the record. The findings suggest that candor cannot erase significant harm to a client or the judicial process, but it can prevent an already serious mistake from becoming a separate problem of professional responsibility.