XPost: free.tampon.tim.walz, mn.politics, sac.politics   
   XPost: talk.politics.guns   
   From: kamala-law@laugh.laugh.laugh   
      
   A federal district judge issued a harsh rebuke and tossed out the   
   testimony of a Stanford misinformation expert who submitted a court   
   document, under penalty of perjury, containing misinformation in a   
   Minnesota election law case.   
      
   Jeff Hancock, who specializes in “research on how people use deception   
   with technology,” was retained by the office of Attorney General Keith   
   Ellison to submit expert testimony defending Minnesota’s new law banning   
   election deepfakes, which was signed in 2023 and updated the following   
   year.   
      
   After Hancock filed written testimony last November, attorneys for   
   plaintiffs Rep. Mary Franson, R-Alexandria, and YouTuber Christopher Kohls   
   noticed that the document contained several citations to academic articles   
   that do not exist.   
      
   The plaintiffs moved to have the testimony thrown out, and Hancock   
   subsequently filed a document admitting he used a version of ChatGPT to   
   draft the testimony, which included the non-existent citations, known   
   among AI researchers as “AI hallucinations.” The Attorney General’s Office   
   argued Hancock should be allowed to file an amended declaration containing   
   correct, non-hallucinated citations.   
      
   But in a ruling dated Jan. 10, U.S. District Judge Laura Provinzino   
   strongly disagreed.   
      
   Hancock’s citation of fake sources “shatters his credibility with this   
   Court,” Provinzino wrote. While acknowledging that artificial intelligence   
   software may have valid uses in legal settings, she concluded that “when   
   attorneys and experts abdicate their independent judgment and critical   
   thinking skills in favor of ready-made, AI-generated answers, the quality   
   of our legal profession and the Court’s decisional process suffer.”   
      
   The judge makes repeated note of the fact that Hancock submitted his   
   original document under penalty of perjury. “Signing a declaration under   
   penalty of perjury is not a mere formality,” she wrote, but is rather an   
   acknowledgement of the “gravity of the undertaking” and a mechanism for   
   ensuring “truthtelling and reliability” as well as trust.   
      
   “That trust was broken here,” she added. “Given that the Hancock   
   Declaration’s errors undermine its competence and credibility, the Court   
   will exclude consideration of Professor Hancock’s expert testimony in   
   deciding Plaintiffs’ preliminary-injunction motion.”   
      
   Provinzino also reminded the Attorney General’s Office they have a   
   responsibility to “validate the truth and legal reasonableness of the   
   papers filed,” and suggested that in the future they should ask witnesses   
   whether or not they used AI to produce any of their material.   
      
   Hancock is billing the Attorney General’s office $600 an hour for his   
   services, according to a copy of the contract obtained by the Reformer   
   under a Data Practices Act request, with billing capped at $49,000.   
      
   The Attorney General’s Office did not provide information on how much had   
   been paid out under that contract so far, or whether the office knew in   
   advance that Hancock would be using AI to draft his declaration.   
      
   “Professor Hancock, a credentialed expert on the dangers of AI and   
   misinformation, has fallen victim to the siren call of relying too heavily   
   on AI — in a case that revolves around the dangers of AI, no less,”   
   Provinzino wrote. “The irony.”   
      
   https://minnesotareformer.com/2025/01/14/judge-rebukes-stanford-   
   misinformation-expert-for-using-chatgpt-to-draft-testimony/   
      
   --- SoupGate-DOS v1.05   
    * Origin: you cannot sedate... all the things you hate (1:229/2)