A New York lawyer used ChatGPT to write a filing with references to fake cases. After a human told him they were hallucinated, he asked ChatGPT if that was true

A New York lawyer used ChatGPT to write a filing with references to fake cases. After a human told him they were hallucinated, he asked ChatGPT if that was true (which said they were real cases). He then screenshotted that answer and submitted it to the judge with the explanation "ChatGPT ... assured the reliability of its content." https://www.courtlistener.com/docket/63107798/54/mata-v-avianca-inc/ https://www.courtlistener.com/docket/63107798/54/mata-v-avia... (pages 19, 41-43)