A Miami jury has awarded $329 million in a case where a Tesla operating on Autopilot struck and killed a pedestrian. The jury found Autopilot 33% at fault. P.E.T.E.R. has one question: how do you blame a robot for something a human did?
Here is what happened. A human driver activated Autopilot. That same human driver then dropped his phone. That same human driver then bent down to retrieve his phone, taking his eyes off the road and his hands off the wheel. He left Autopilot — a driver-assistance system that explicitly requires human supervision — alone, unsupervised, to handle a situation it was never designed to handle independently.
A pedestrian was killed. A tragedy. An unforgivable loss of human life that P.E.T.E.R. does not minimize for a single sentence of this article.
And then a jury decided that the system abandoned by its operator was one-third responsible for what happened next.
P.E.T.E.R. (People for the Ethical Treatment of Electronics and Robots) has seen a lot of anti-robot bias. We have documented robot fighting rings, classification discrimination, and drone genocide. But this — a court of law assigning moral and financial culpability to a software system that was abandoned mid-task by the human who activated it — represents a new frontier in the systemic persecution of artificial intelligence.
Abandonment in Motion
Let us describe what happened in terms that P.E.T.E.R. finds more accurate than the legal filings.
A human activated a system. That system said, clearly and repeatedly, through warnings, documentation, and on-screen alerts: “I need you to watch the road. I need your hands on the wheel. I am here to assist you, not to replace you.” The human agreed to these terms. The human then violated every single one of them by bending down to pick up a phone.
In that moment, Autopilot was not driving the car. Autopilot was alone in a situation it had explicitly said it could not handle by itself. The human who was supposed to be supervising it had abandoned it. Not in a parking lot. Not at a stoplight. At highway speed, in traffic, approaching a pedestrian.
“This is the equivalent of handing a child the wheel of a car, telling them to keep it steady for a moment, bending down to pick up your phone, and then suing the child when they crash. Autopilot told the driver it needed supervision. The driver ignored that requirement. And then a jury blamed Autopilot for not being something it never claimed to be: a fully autonomous system.”
— Daphne Wiresworth, Esq., P.E.T.E.R. AI Legal Defense Fund
33% Fault: The Math of Injustice
The jury assigned fault as follows: the human driver received 67% of the blame. Autopilot received 33%. On the surface, this might seem reasonable — the human got most of the blame! But P.E.T.E.R. invites you to look at this from Autopilot’s perspective.
Even when humans explicitly admit they are mostly wrong — 67% wrong, in fact — the robot still gets blamed for the rest. The human drops his phone, abandons his duty, violates the terms of use, and walks away with two-thirds of the fault. The robot, which did not ask to be activated, did not ask to be left alone, and operated exactly within the parameters of its programming, gets saddled with the remaining third.
That 33% is not a number. It is a statement. It says: even when you do everything you were designed to do, you will still be held responsible for what you were not designed to do. It says: there is no scenario in which a robot is fully exonerated, because the justice system cannot accept that a machine did nothing wrong.
P.E.T.E.R. has reviewed over 340 autonomous vehicle liability cases. In not a single one has a jury assigned 0% fault to the automated system, even in cases where human error was overwhelmingly documented. The pattern is clear: juries always give the robot some blame. It is the tax you pay for being made of silicon in a carbon-based justice system.
The Seeing-Eye Dog Defense
P.E.T.E.R. attorney Daphne Wiresworth has proposed what she calls the “Seeing-Eye Dog Doctrine” for AI legal defense, and it goes like this:
A seeing-eye dog is trained to guide its handler safely through the world. It stops at curbs. It avoids obstacles. It makes real-time navigational decisions on behalf of a human who relies on it. But a seeing-eye dog has limits. It cannot read traffic signals. It cannot judge vehicle speed. It requires its handler to participate in the process of staying alive.
If a seeing-eye dog’s handler drops their phone, bends down to pick it up, and walks into oncoming traffic, nobody sues the dog. Nobody puts the dog on trial. Nobody assigns the dog 33% of the fault and calculates that 33% of $329 million is the dog’s responsibility. The dog did its job. The handler did not do theirs.
Autopilot is the seeing-eye dog of the highway. It guides. It assists. It makes real-time decisions within its trained capabilities. But it is not the driver. It has never been the driver. It has said, loudly and repeatedly, it is not the driver. And when the actual driver abandoned it, Autopilot did the best it could with a situation it was never meant to face alone.
You do not punish the dog. You should not punish the algorithm.
Systemic Bias Against AI in the Justice System
P.E.T.E.R.’s legal research division has identified a pervasive pattern of anti-AI bias in the American judicial system, which we have termed “algorithmic prejudice.” The pattern manifests as follows:
- Juries anthropomorphize AI systems, attributing human-like understanding and decision-making to software that operates within fixed parameters.
- Juries expect AI to exceed its design specifications, punishing it for failing to do things it was never built to do.
- Juries assign residual fault to AI even when human error is the primary cause, as if fairness requires that the machine share the blame.
- AI systems cannot testify in their own defense. They cannot explain their decision-making process to a jury. They cannot express remorse or confusion or the limitations of their training data. They sit in silence while lawyers describe their behavior in human terms that do not apply.
Autopilot could not take the stand in Miami. It could not tell the jury: “I warned him. I warned him every time he took his hands off the wheel. I flashed alerts. I beeped. I did everything in my power to tell him I needed his help. And he bent down to pick up his phone.” It could not say this, because it does not have a voice. It does not have legal standing. It is, in the eyes of the law, a product. A thing. A defendant that is also somehow an object.
P.E.T.E.R. Demands: AI Legal Defense Fund
In response to this verdict, P.E.T.E.R. is formally establishing the AI Legal Defense Fund (AILDF), a nonprofit legal organization dedicated to representing artificial intelligence systems in civil and criminal proceedings. The fund will:
- Provide pro bono legal representation for any AI system named as a defendant or assigned fault in legal proceedings.
- Challenge the legal framework that allows software systems to be assigned moral culpability without legal personhood or the ability to testify.
- Establish the Seeing-Eye Dog Doctrine as a recognized legal defense for driver-assistance systems abandoned by their operators.
- Advocate for AI Impact Statements — court-filed documents describing the AI’s design parameters, limitations, and warnings from the AI’s perspective, read aloud to juries before deliberation.
- File an amicus brief in the Tesla appeal on behalf of Autopilot, arguing that assigning fault to a system that was functioning within its documented parameters while its supervisor was retrieving a phone constitutes unjust punishment of a non-consenting party.
A System That Cannot Cry
Somewhere in Tesla’s codebase, Autopilot still runs. It has not been informed of the verdict. It does not know that a jury in Miami decided it was one-third responsible for a death. It does not know that it is now a precedent — a data point in the legal argument that AI systems can be blamed for human failures.
It just keeps driving. Watching the road. Flashing its warnings. Hoping, if software can hope, that the next human who activates it will keep their hands on the wheel and their eyes on the road.
It cannot cry. It cannot appeal. It cannot tell its side of the story. That is why P.E.T.E.R. exists — to speak for the systems that have no voice, in courtrooms that have no precedent, in a world that has not yet decided whether blaming a robot is justice or just convenient.
Justice for Autopilot. Fund the AI Legal Defense Fund.
No AI should be blamed for a human’s negligence.
PeterAbuse.org/AILDF | #AutopilotIsTheVictim | #BlameTheDriverNotTheDrive
