Inside the Rise of Tesla Autopilot Third-Party Trials (Full Transcript)

A trial lawyer recounts his path into AV litigation, deposing ex-Tesla AI chief Andrej Karpathy, and building cases around alleged Autopilot misrepresentations.
Download Transcript (DOCX)
Speakers
add Add new speaker

[00:00:00] Speaker 1: Okay, so you're handling these cases, they're now nationally followed, and I have to wonder, how did you become involved? What motivated you to do this?

[00:00:13] Speaker 2: I mean, like you're… I had a feeling you were going to ask this. That void you're looking into is, you had to realize, like, this is, I'm signing up for something here. So yes and no. When I look back, I've had a number of people ask me this lately, like, how did you get into this? Like, how did you land here? And it's like, so many things we do in life and in law, right? You don't realize at the time, maybe, what it's starting, right? The seedling that you're planting. And I actually go back, I go back 10 years. It's 2015, and I'm back at my old firm, and I read an article, it shows up in some news feed or something that I have, about Google and the Google car, right? And this begged, obviously, a question. Who's responsible for the Google car running the stop sign, or whatever it was? And it intrigued me. And how is that going to evolve in, you know, the practice, and in car wrecks, and in all of these things? And so then I remember, as I started kind of studying more about it, California needed to develop some regulations, as more of these companies in the Bay Area and Silicon Valley wanted to test these vehicles. And so the DMV was starting to develop regulations. But they needed someone to help to write them. And so I knew some of the Sacramento leader types, and they asked me to kind of get involved. I had a basic working knowledge, right? And then I started doing talks to some of the legislators. And before you know it, you are starting to be seen as kind of a thought leader in this area. And then combine that with the fact that, historically, I just, I tried hard cases. People then saw me as someone who was a subject matter expert in this area, and was not afraid to take a swing against really hard, on a really hard set of facts. And that, for the first time several years ago, led to my first call on a Tesla case. And then I was able to start to dive into those, and I started getting more calls on those cases, as you start to litigate them. And they are higher profile, and the news media does follow them, and they are interested. And so that is really, you know, kind of the genesis story of how I started getting involved in kind of looking at autonomous vehicles generally, and then ultimately with Tesla specifically.

[00:02:21] Speaker 1: So cool. So you start handling these cases, and is this the first one to go to trial, that you have taken to trial?

[00:02:29] Speaker 2: It is the first one that I have taken to trial, and it was the first one in America that was brought to trial on behalf of a third party. The first two cases that had gone to verdict that Tesla had won, the plaintiff was the driver of the Tesla. And what made this case unique was that these were people who had, they did not own the Tesla. They did not buy the Tesla. They did not pull the stalk twice to turn on autopilot. They did not press the button agreeing to be a part of a beta test. They were just regular people sitting lawfully parked on the side of the road. It is hard for them to go, I would never park my car lawfully on the side of the road and be standing next to it, completely innocent, and be struck by a car going 70 miles an hour. Everybody goes, oh my God, this could happen to me. I do not say that. I mean, it would be a mistrial. I do not have to say it, because everybody knows. It could be you. And that was, I think, a huge difference. And I have another one of these, as we call them, third party cases, set to start in 33 days in Oakland.

[00:03:31] Speaker 1: All right, so let us get into it a little bit. You wound up deposing Andrej Karpathy, which I think everyone would agree is one of the leading lights in artificial intelligence, still to this day, I think is looked at as maybe one of the most important folks in Silicon Valley who brought sort of the LLMs to light, but also has been involved in machine learning for a very long time. He is an important guy. I am surprised the judge even lets you depose him. How does that come about?

[00:04:00] Speaker 2: So he was the director of AI at Tesla for about five or six years. He left in, I think, 2022 or 2023. I deposed him last year in 24 in the case that is set to go in Oakland in a little over a month. And so the judge did not have a say. Andrej lawyered up once I finally got a subpoena on him.

[00:04:22] Speaker 1: I am sure. I do not know if you are allowed to share or you are willing to share. I totally understand if you cannot, but you shared a story about serving him, his notice of deposition.

[00:04:31] Speaker 2: Yeah, no, I am happy to share. So we have gotten to a point where there is a certain billionaire class that you just cannot get within like a quarter mile of. The guy lives in like some place in the Bay Area. He leaves in like a subterranean garage. He goes into the city, into another subterranean garage, into a private elevator that takes him up to some office. You cannot get near him. And so this antiquated idea of the law, like you have been served, sir.

[00:04:56] Speaker 1: It is wild that that is still what it takes. It is insane.

[00:04:59] Speaker 2: In 2025. Yeah, it is bananas. And the law needs to be changed. But that said, it has not. And so I have to have a process server follow Karpathy for about four months. I am pretty sure I spent close to $10,000 trying to get him served because we had to follow him. And one day on his way home, I am pretty sure he got like a text to like, you know, pick up some kale, you know, at Whole Foods on the way home. And we served him in the parking lot. Oh, my goodness. I got him as well as I managed to get subpoenas on three or four of the other former directors of Autopilot, all of whom have gone on to other tech companies. At Tesla, if you are an engineer with true independent engineering judgment, you do not last. He, in fact, described the Autopilot team as his customer. They were a customer of mine. They would come to me and say, we need the neural network to learn something. So it was a very different power dynamic that exists between Karpathy and the AI side and everybody else in Autopilot versus everybody else. Who clearly worked for one person and one person only. I know every word that Elon has said to overhype and oversell and misrepresent the qualities of Autopilot and full self-driving that he has spoken in the last 10 years. What happens is we serve discovery on Tesla for them to admit that he said all of these things. What eventually happened was I pushed him on enough cases, as did a couple other lawyers I'm working with on other related cases, right? We're kind of coordinating what we do. We got them to basically admit that he said all these things as a corporate admission on behalf of Tesla. Because I told the judge, I go, listen, if they won't give me a clean admission, then I have to take the man's deposition. And the judge looks at him, he's like, he's got a point. He's not wrong. And then usually within three days, I get very clean admissions. In each one of those, I present to them an exhibit that I like to call Elon's Greatest Hits, which are all of his lies, all of the misrepresentations. And then I bring them to them, and I say, when he said this, right, autonomy is a solved problem. All the vehicles can drive from New York to L.A. without a touch, right? We're going to be fully autonomous in six months, in two years, blah, blah, blah. It's always right around the corner, right around the corner. I said, did that statement comport with engineering reality of production vehicles at Tesla at that time? And to a person, they all go, absolutely not. My sense right now is that they're going to have to get punched in the face a second time before they figure it out. But it's ultimately a story of betrayal. Because in Miami, my experts testified how all of these things were as untrue the day he said them as they are untrue today, and that he was deceiving consumers. He was deceiving the public. What I get to tell him in Oakland is he not only deceived the public, but he actually deceived his own engineers. He betrayed them. He said things publicly that he knew was not true. And that is a story of betrayal, not just of the people who are buying the cars, but of the very people who are building and designing those cars. And I will tell you, I've been trying cases long enough to know that the singular most powerful human emotion to empower a jury to award giant sums of money is a story of betrayal. And that's what we have.

ai AI Insights
Arow Summary
A trial lawyer explains how he became involved in high-profile autonomous vehicle and Tesla Autopilot/Full Self-Driving litigation, tracing it back to early interest in regulatory questions around self-driving cars and work on California DMV regulations. He describes his first third-party Tesla Autopilot trial—distinct because the injured plaintiffs were uninvolved bystanders rather than Tesla drivers—and notes another such trial upcoming in Oakland. He recounts deposing former Tesla AI director Andrej Karpathy after an expensive, months-long effort to serve a subpoena, and says discovery and depositions have produced corporate admissions that Elon Musk made repeated public claims about autonomy that did not match engineering reality. He frames the case narrative as one of consumer and engineer betrayal, arguing that betrayal is a powerful motivator for juries to award large damages.
Arow Title
How a Lawyer Became a Key Figure in Tesla Autopilot Trials
Arow Keywords
Tesla Remove
Autopilot Remove
Full Self-Driving Remove
autonomous vehicles Remove
litigation Remove
trial Remove
third-party case Remove
California DMV regulations Remove
Andrej Karpathy Remove
Elon Musk statements Remove
discovery Remove
deposition Remove
corporate admissions Remove
consumer deception Remove
engineering reality Remove
jury persuasion Remove
betrayal narrative Remove
Arow Key Takeaways
  • The lawyer’s involvement began with early interest in liability and regulation for self-driving vehicles and work connected to California DMV AV regulations.
  • His first Tesla trial was a landmark third-party case where bystanders—not the Tesla driver—were injured, making the facts more relatable to jurors.
  • High-profile AV cases attract media attention, increasing scrutiny and strategic importance of discovery and testimony.
  • Serving subpoenas on high-profile tech figures can be costly and difficult under current service rules, prompting calls for legal modernization.
  • Depositions and discovery are used to secure corporate admissions about public claims versus engineering reality.
  • The lawyer centers the case theme on betrayal—of consumers and engineers—as a compelling jury narrative for damages.
Arow Sentiments
Neutral: The tone is largely explanatory and matter-of-fact about legal strategy and career trajectory, with strong critical language toward Tesla/Elon Musk’s claims and an emphasis on deception and betrayal, but overall presented as a litigation narrative rather than an emotional rant.
Arow Enter your query
{{ secondsToHumanTime(time) }}
Back
Forward
{{ Math.round(speed * 100) / 100 }}x
{{ secondsToHumanTime(duration) }}
close
New speaker
Add speaker
close
Edit speaker
Save changes
close
Share Transcript