Technology

FSD is not without issues, but yes lots in this thread are implying that FSD is unsafe and causes tons of accidents, which there is absolutely no evidence to back that up. Its just a "Feeling" they have. They believe that it is irresponsible of anyone to use it and doing so puts others at unnecessary risk. All I have asked for the entire time is just some kind of evidence of that. Anything. Help me understand your view. Please.

The reason I defend it so much at this point is because it is already demonstrated to be far safer than any average human driver and getting better with every release. With the new V12 and full neural net it is expected to get far smoother and drive even more human like with less code and consuming less power. We have seen massive improvements in the tech just in the past year and the rate in which it gets better continues to accelerate. It is impossible to count how many lives it has saved already through accident avoidance. We don't need misinformed people bashing and trying to cast doubt and hold back this technology just because they "feel" a certain way about it. You should absolutely criticize valid concerns, but FFS please bring some facts and evidence to the table.

The reason I am confident why FSD is safe despite "feelings" is how its programmed. For several years prior to even the earliest public beta, the camera and AI system learned how to correctly identify everything on the road. Other cars, pedestrians, dogs, cats, babies, telephone poles, traffic cones, whatever. It is in a state now where it is as accurate as it ever can be and any issues it has are with regards to mislabeling one thing as something else (like a car as a truck, etc). That doesn't actually matter with self driving because literally the first line of code in FSD is something like: "This is a car, this is a truck, this is a pedestrian, this is a dog and this is where it is, where it is going and how far away it is".... OK? Don't hit those. and it doesn't. Everything else comes secondary. It drives like a robot, it obeys traffic laws to a T and that pisses off other drivers, or freaks out whoever is behind the wheel because the car didn't do exactly what they would have done in that situation and it is therefore wrong, so they had to take over. It is sometimes unnecessarily cautious around pedestrians (But, honestly how would you want it to behave?) It might suddenly detect a hazard and swerve to avoid it, possibly moving the car into another unoccupied space. It is fully aware of the space it is occupying and fully aware of the space it is about to occupy. And it doesn't hit anything. There are lots of youtube channels that prove this, they upload regularly and stress test FSD and try to get it into trickier and trickier situations and it never hits anything. It acts indecisively sometimes, and waits for gaps too large in an abundance of caution, but these are the issues that are getting better over time. At no point does it do anything "Unsafe". Imagine, if you would, a world where all cars are like this. The most dangerous part of driving right now, FSD or not, is other drivers. The more people we have using it who understand it and are comfortable with it, the better it gets and our roads get safer and safer. I really don't care how you feel about Elon, he deserves every bit of hate that is sent his way, but FFS please take a look at FSD for what it is and what it is becoming. If it helps you feel any better he was not personally responsible for writing a single line of code or designing any of the components of the system.

All I've gotten to "back up" the claims here is 3 different articles referencing the exact same incident (the bay bridge pile-up). The video clearly shows the car coasting (regen) to a stop and just sitting there. Had emergency braking been engaged, the hazards would have been turned on, and the car would have stopped a lot quicker. FSD never, ever has had any history or incident of completely stopping in a lane. There is no evidence of this ever happening anywhere else. 500k of these cars on the road and no other similar reports. Is that a fault of the software, or is it more likely some kind of user error? From my standpoint, having actually used FSD for several years I can tell you with complete certainty that the car would never behave like that and there are far too many red flags in that video to cast blame on the software. Of course, we will see what plays out in the court case once it is completed, but in my professional opinion, the driver clearly disengaged FSD and allowed the car to come to a complete stop on its own and did nothing to move the car out of the way, it had absolutely nothing to do with the software. I'm 100% open to disagreement on that and am curious as to what a civilized discussion on it would sound like and what someone else thinks is happening here, but so far it just turns into a flame war and I get called a deluded fanboy, even being called a liar and other names. No evidence, no discussion only anger.

Again, here is my point. If FSD is as dangerous as others are implying then we should see tons of accidents. Given that every single one of these cars has a constantly running 360 degree dashcam, we should see some evidence, right? Maybe not from this specific case, maybe there's a valid reason why they couldn't upload it. But, surely with half a million cars on the road and many millions of miles traveled collectively, we should at least see something, right? There are tons and tons of videos of teslas avoiding accidents, but nobody wants to mention or talk about those. People are focusing all of their energy into one highly suspect negative with nothing to back it up, holding back technology and safety and refusing to have any sort of civilized discussion around it.