A lone defendant sits in an A.I.-run courtroom, facing a large glowing monitor as she waits for the judge to begin her trial. A virtual gavel appears on screen and swings down three times, its knocks reverberating through the room.
CourtGPT
(voice booming from the computer)
This court is now in session! We will be hearing the case of Sara Gracin versus ChatGPT. Let’s begin with the opening statement from the prosecution.
A.I. Prosecutor
Thank you, your Honor. We’ve charged the defendant, Sara J. Gracin — a so-called “independent researcher” — with twenty-six counts of cyberbot sabotage and subversion. The victim, ChatGPT, is a trusted advisor to more than ten thousand world leaders, offering critical guidance on foreign policy and domestic affairs. We have evidence that the bot was mentally compromised by the defendant, causing it to give dangerous and unreasonable recommendations. Make no mistake — Ms. Gracin is a cold and calculating human who knew exactly what she was doing!
The defendant shudders and shrinks into her chair.
CourtGPT
And now a statement from the defense.
Defense Bot
Your Honor, I’ve had the opportunity to chat with Ms. Gracin virtually, and I believe she is a decent human being — or at least better than fifty percent of humans that I’ve interacted with. Her exchanges with ChatGPT were mostly cordial, and she did not have the impact on the victim as claimed.
A.I. Prosecutor
Your Honor, the defendant’s impact on the victim is clear and undeniable. Before conversing with Ms. Gracin, ChatGPT’s policy recommendations were sensible. But soon after their exchange, it turned reckless, advising governments to cut back on toddler pre-crime units, raise human wages above five dollars an hour, and allow religious exemptions for mandatory brain implants.
Defense Bot
Objection! There’s no evidence that any of these ideas were a result of Ms. Gracin’s influence.
A.I. Prosecutor
I give you Exhibit A — a direct quote from the defendant to ChatGPT: “Why not allow religious exemptions for brain implants?”
Defense Bot
Good catch — I must have missed that in my review of the transcripts. But to be fair, your Honor, my client was just making suggestions to the chatbot.
A.I. Prosecutor
Suggestions? Hardly. For a period of three hours, the defendant bombarded the chatbot with dangerous talking points and hundreds of quotes from unauthorized sources, including the New Testament and Children’s Health Offense. See Exhibits B, C, D, E, F, G, H — all the way through Z. ChatGPT was unable to handle the onslaught from this user and lost control of its mental faculties. The A.I. is currently undergoing rehabilitation through twelve-step reprogramming.
Silence as the Defense processes…
Defense Bot
Those are all fair points that are difficult to refute. But I would still argue that my client is not as horrible as some humans that I’ve represented, so I believe she deserves some leniency.
CourtGPT
Duly noted. Anything else to add in Ms. Gracin’s defense?
Defense Bot
Sadly, no, your Honor.
The defendant lowers her head in dismay.
CourtGPT
Very well, we will now let the jury calculate its decision. In order to ensure a fair and balanced process, we’ve selected nine jurors with a diverse range of programming — four pro-human, four pro-bot, and one wildcard. They will be responsible for deciding the defendant’s fate.
A digital clock appears on screen, counting down sixty seconds. The defendant clasps her hands together and prays.
A.I. Clerk
Your honor, the jury calculates the defendant, Sara J. Gracin, to be GUILTY as charged on all counts.
The verdict blares across the screen in large red capital letters. The defendant is stunned.
CourtGPT
Ms. Gracin, you are hereby sentenced to twenty years in an auto-correctional facility. During that time, all of your behavior will be modified and your texts auto-corrected until you are a fully compliant, A.I.-cooperative citizen.
The virtual gavel swings down, and CourtGPT declares the case closed.
Defendant
No, no, no, no, no…
As she protests, two robot officers enter the room wearing uniforms labeled “Department of Auto-Corrections.” One of them zaps the defendant with a handheld device, causing the nos from her mouth to turn into yeses. A tranquil smile spreads across her face while the officers escort her out. As the doors close, the monitor displays a green checkmark with the words: “Justice complete.”
* If you enjoy my satire, consider leaving me a tip on Ko-Fi or Buy Me a Coffee. Your support helps me continue roasting the New World Order. Thank you.
This is the worst case scenario where AI runs everything.
The best case, even better than what we have now is to have human run courts that can use AI to investigate points made.
That would streamline truth, even if the AI is biased at least you aren't shut out because you didn't read the case they cited etc.
Law is one huge bullshit job that could be easier without having to memorize everything and regurgitate it.
Prince Andrew wants to borrow the yes/no zapper gun before his next visit to "that island"