The Drunk Driver Argument for FSD Is Real. So Is the Legal Problem.
```html
A man in Vacaville, California passed out drunk behind the wheel of his Model Y, and the car drove itself through busy streets without hitting anything. That's a remarkable sentence. And depending on who you ask, it's either a compelling argument for FSD or a cautionary tale about letting technology outpace common sense (and law).
I've been thinking about this incident in the context of something Elon Musk said to investors in December 2025: that FSD v14.2.1 is good enough that texting while driving "depending on context of surrounding traffic" is something the system can handle. He went further, arguing that a driver turning off autopilot to check texts while steering with their knees is "significantly less safe" than just leaving FSD engaged. He called FSD "kind of the killer app" because of that safety gap.
He's Not Wrong About the Safety Math
The uncomfortable truth here is that Musk's underlying argument has some validity. If FSD in supervision-required mode is genuinely safer than a distracted human driver, then the math favors leaving it on. The Vacaville incident is an extreme example of that logic playing out in real life. A passed-out drunk is essentially a worst-case distracted driver, and the car handled it.
But there's a difference between "the technology worked" and "this is legal or advisable." California law is clear: FSD is permitted, but the driver has to be conscious, alert, and sober while operating the vehicle. The fact that the autopilot saved the day doesn't change the man's legal exposure. The law doesn't have a "but the car did fine" exception.
The Gap Between What FSD Can Do and What You're Allowed to Do With It
This is the tension that's going to define FSD policy debates for the next few years. The technology is outpacing the legal framework, and Musk's public statements about texting while FSD is engaged aren't helping clarify things. "Depending on context" is not a legal standard. It's not something a cop or a judge can work with. And it puts drivers in an awkward position where the CEO of the company is implying it's fine to look at your phone, while state law says you need to be alert and in control.
I'm not going to tell you what to do on your own commute. But I'd note that "Elon said it was okay" has never been a successful defense in traffic court.
What the Vacaville Incident Actually Tells Us
The drunk driver story is genuinely striking because it shows FSD operating in a scenario it was never designed for. No driver monitoring, no input, presumably no intervention available. And it worked. That's worth acknowledging.
But the reason we have driver-presence requirements isn't just bureaucratic caution. It's because FSD still fails. Not always, not even often, but it fails. The Vacaville driver got lucky. Someone else in the same situation might not. The technology isn't at the point where unconscious supervision is reasonable, even if this one incident came out okay.
Musk's point about distracted driving is legitimate. The texting-while-steering-with-your-knees comparison is real and it's dangerous. If FSD can make that scenario safer, that matters. But "safer than the worst version of human driving" is a low bar, and it's not the same as "safe enough to remove legal responsibility from the driver entirely."
Where This Leaves Owners
If you're using FSD regularly, the practical guidance hasn't changed: California (and most states) require you to be awake and paying attention. Musk's comments about texting in low-traffic contexts aren't legal cover, and the Vacaville story isn't a template to follow. It's an anecdote about a system doing something impressive under circumstances that shouldn't happen.
The capability is genuinely getting better. v14.2.1 is reportedly a meaningful jump in handling complex scenarios. But the legal framework is going to lag the technology for a while, and in that gap, the liability sits with you, not with Tesla.
Stay awake. It's still your license.
Source: Teslarati
```