Upvoted because I think it's an important topic, but this take causes me to question the motive for the article... which ironically is my big concern with using LLMs to write stuff generally (the unconscious censoring / proctoring of voice and viewpoint):

  That means that if an officer is caught lying on the stand – as shown by a
  contradiction between their courtroom testimony and their earlier police
  report – they could point to the contradictory parts of their report and say,
  “the AI wrote that.”

IANAL but if they signed off on it then presumably they own it. Same as if it was Microsoft Dog, an intern, whatever. If they said "the AI shat it" then I'd ask "what parts did you find unacceptable and edit?" and then expect we'd get the juicy stuff hallucinations or "I don't recall". Did they write this, or are they testifying to the veracity of hearsay?

From what I've seen reports written by / for lawyers / jurists / judges already "pull" to a voice and viewpoint; I'll leave it there.