Artificial intelligence detection company GPTZero analyzed all 4,841 papers accepted at the NeurIPS conference and found 100 instances of false citations in 51 papers. Although these false citations are statistically insignificant compared to tens of thousands of citations, this discovery highlights the challenge of maintaining academic rigor amid an AI-driven "submission tsunami." NeurIPS pointed out that incorrect citations do not necessarily invalidate research, but fabricated citations diminish their value as key academic indicators. This incident has also sparked widespread concerns about the accuracy of artificial intelligence, even among top experts who are worried about it.