In March, the Arizona Supreme Court launched two AI-generated avatars named Victoria and Daniel: thats the pair above. These AI, non-existant personas deliver news of judicial rulings and opinions in the state via YouTube videos. Jerome Dewald, a 74-year-old plaintiff was inspired to say, “Hold my beer!”
Dewald created an AI-generated video avatar to deliver his argument via Zoom in court. Five New York State judges at the New York State Supreme Court Appellate Division’s First Judicial Department were anticipating his pro se presentation in an employment case on March 26, but instead of the elderly litigant they saw a young man in a button-down shirt and sweater.
“May it please the court,” said the un-named avatar. “I come here today a humble pro se before a panel of five distinguished justices.” Justice Sallie Manzanet-Daniels, interrupted the presentation before the avatar (the avatar’s pronouns were “it” and “it”) could speak another word , saying “Okay, hold on. Is that counsel for the case?” After Dewald confirmed that he had generated the non-lawyer non-person using AI, Manzanet-Daniels ordered the video to be turned off.
“I don’t appreciate being misled,” she said, for Dewald had not revealed to the court that he would be using a bot-created “lawyer” to present his case. The judges allowed the pro se human being to make his argument himself, and Dewald later sent an apologetic letter to the court explaining that he couldn’t afford a lawyer and thought an AI “attorney” would be more presentable and persuasive.
This, my friends, is only the beginning (that’s the first line in the finale of one of my favorite musicals, Anthony Newley’s “The Roar of the Greasepaint, the Smell of the Crowd”). Nobody knows how much and to what extent artificial intelligence will invade our justice system, and if you want to give odds on AI judges eventually hearing cases, I’ll take that bet.
In June 2023, the Eleventh Judicial Circuit of Florida introduced an AI digital chatbot named Sandi, which offers assistance for visitors to the Miami-Dade Courts website in English or Spanish. As Margot Channing warned,
[Note: This is funny. WordPress’s bot is so Trump-obsessed that it told me this post should be tagged “Donald Trump” even though nothing in it is even vaguely related to the President.]
Ethics Alarms 
Well, since we have an AI theme (and comments are sparse today), I was curious how a current AI model would respond to this post and asked Deepseek the following: “Comment as if you were a reader of this blog post and commenting in the comment section.”
Here is the response from Deepseek.
Here’s what my buddy, Grok 3 (the X AI) had to say about it:
“Interesting take on AI’s role in the legal system! Jerome Dewald’s creative use of an AI avatar to present his case, while ethically questionable due to the lack of transparency, highlights a broader issue: access to justice. His apology and explanation about affordability underscore how cost barriers push people toward unconventional solutions. AI tools like Arizona’s Victoria and Daniel or Florida’s Sandi show promise in making legal processes more accessible, but they also raise valid concerns about authenticity and trust in the courtroom. I’d wager we’re far from AI judges, but the line between helpful tech and deceptive use needs careful policing. Thanks for sparking this discussion!“
Asking it “Can you use a more humorous tone, and address the ethics issues a bit more?” got this:
“Oh, Jerome Dewald, you sly dog! Deciding to yeet an AI-generated Chad in a sweater to argue your case via Zoom? That’s the kind of “Hold my beer” energy I can respect—but, alas, the court was not here for it. Justice Manzanet-Daniels shut that AI attorney down faster than you can say “objection, your honor!” And rightfully so—springing a digital doppelgänger on the judges without a heads-up is less “innovative” and more “ethically oops.” Misleading the court, even unintentionally, risks undermining trust in the whole justice system. Nobody wants a trial where you’re left wondering if the lawyer is human or just a fancy chatbot with a law degree from the Matrix.
Still, I get it. Legal fees are scarier than a horror movie jump-scare, and Dewald’s DIY AI stunt screams desperation for affordable representation. Arizona’s AI avatars, Victoria and Daniel, and Florida’s chatbot Sandi are at least upfront about their silicon souls, helping folks navigate the system without pulling a fast one. But here’s the ethical tightrope: AI can democratize access to justice, sure, but if we’re not crystal clear about who—or what—is speaking, we’re flirting with a courtroom circus where avatars argue and judges wonder if they’re ruling over humans or holograms. Let’s keep AI as a trusty sidekick, not a sneaky stand-in, okay? Loving this wild ride of a topic!”