Tesla Full Self Driving Thread
Lexus Test Driver
Joined: Mar 2013
Posts: 5,441
Likes: 946
From: IL
Thread Starter
Lexus Champion
Joined: Oct 2019
Posts: 19,428
Likes: 3,656
From: California
I can say this thread has definitely taught me a lot about people. Don't want to say anymore on that front.
Officially now a monthly subscriber. With the money I'm saving on gas (a lot), it's worth letting the car do all the work for my daily commute
Capability, but until all or most cars on the road are being driven autonomously? LONG way off.
I'm not overstating my ability as a driver. I said I was an "above average driver", and I am. I didn't say I was a stunt driver, or an F1 driver, or a secret service driver lol. I have made becoming a great driver a focus in my life and gave spent time and energy being trained and practicing etc, I think anybody would say that makes me "above average".
I'm not overstating my ability as a driver. I said I was an "above average driver", and I am. I didn't say I was a stunt driver, or an F1 driver, or a secret service driver lol. I have made becoming a great driver a focus in my life and gave spent time and energy being trained and practicing etc, I think anybody would say that makes me "above average".
Its as simple as, you have not had an accident in 30 years. Would you be able to say the same if you had current FSD 30 years ago? Doubt it. If the manufacturer of any self driving software is not wiling to take on the liability then why should I? Tesla doesn't trust their own software but yet still you should?
Isn't the data being used to compare FSD to Humans skewed? I am sure a large subset of accidents that occur from human error is because of distracted driving (on phones), intoxication, less experienced drivers (under 30) and reckless driving. If you don't fit in any of those categories your chances of getting in an accident is drastically reduced.
This s a 2026 Model Y HW4 14.2 getting in an accident. Not sure about the Pro FSD people here, but the chances of this same incident occurring from me driving is right around 0%.
https://tinyurl.com/f3vzn25b
Another example of HW4 14.2 heading towards a barrier at 56 mph
https://tinyurl.com/mux93r37
Another example of HW4 14.2 where FSD turns onto a boat ramp and about to drive into a lake instead of turning into the owners driveway.
https://www.reddit.com/r/TeslaFSD/co...e_into_a_lake/
Tesla isn't even willing to take liability for Summon....this doesn't speak volumes? Tesla doesn't even trust their own software to navigate a parking lot but I should believe it can drive better than me on a highway at 80 MPH?
I know I will get the usual I am a hater or the examples above gets dismissed by Pro FSD people here. However, seeing these simple mistakes happen, ain't no way I am trusting it in complex situations.
Lexus Fanatic
Joined: Oct 2003
Posts: 94,203
Likes: 218
From: Virginia/D.C. suburbs
Well, in this endless Tech-vs-Self-Ability debate, my own experience has been somewhere in the middle. Like Steve, I am an Above-Average-driver for my age (that's not my opinion, but a fact). But, considering my age, that still means some limitations...such as the ability for my eyes to adjust to morons coming at me at night with their bright-lights on. My reflexes, although not those of a 20-year-old, are better than most drivers my age. However, I have found that some tech-features usually work better than I do, such as the Rear-Cross-Traffic alert in spotting obstacles while backing out of partially-blind parking places, and the Blind-Spot-Alert with the yellow-orange warning light in the side-mirrors, which tells you that someone is in your blind-spot when you went to change lanes. Other features, such as with the Lane-Keeping and Self-Centering that is supposed to keep you from drifting out of your lane, don't seem to work very well, particularly if the road has difficult-to-see lane markings in bad weather, road construction, or at night....with those, I myself usually do a much better job.
Last edited by mmarshall; Apr 11, 2026 at 07:27 PM.
Lexus Fanatic
Joined: May 2012
Posts: 67,892
Likes: 3,831
From: Maryland
What do you think would make someone an above average driver then?
FSD is like EVs. There is an evangelism. People feel
personally attacked when others aren’t interested.
Last edited by SW17LS; Apr 11, 2026 at 06:53 PM.
Lead Lap
Joined: May 2010
Posts: 3,866
Likes: 1,468
From: San Diego, CA
I don't get why the pro FSD guys here are giving you a hard time (although I know you don't let it bother you lol).
Its as simple as, you have not had an accident in 30 years. Would you be able to say the same if you had current FSD 30 years ago? Doubt it. If the manufacturer of any self driving software is not wiling to take on the liability then why should I? Tesla doesn't trust their own software but yet still you should?
Its as simple as, you have not had an accident in 30 years. Would you be able to say the same if you had current FSD 30 years ago? Doubt it. If the manufacturer of any self driving software is not wiling to take on the liability then why should I? Tesla doesn't trust their own software but yet still you should?
FSD is a nascent technology, but I already consider it to have easily surpassed the driving ability of the average American. And while the low driving standards continue to decline, FSD and similar systems are continually improving.
Isn't the data being used to compare FSD to Humans skewed? I am sure a large subset of accidents that occur from human error is because of distracted driving (on phones), intoxication, less experienced drivers (under 30) and reckless driving. If you don't fit in any of those categories your chances of getting in an accident is drastically reduced.
Of course there are going to be issues, FSD is a nascent technology. But again, you have the wrong basis for comparison. FSD has logged just under 9.2 billion miles at the time of posting. The statistics tell you that if you were to look at 9.2 billion miles of equivalent human driving, you are going to find more examples of human error than FSD error.
Bottom line - the argument isn't that FSD doesn't do stupid things. The argument is that it will do them less often than a person, and that while still nascent FSD is getting better at a rapid pace while the standard of human driving is gradually declining.
Last edited by swajames; Apr 11, 2026 at 07:24 PM.
Lexus Fanatic
Joined: May 2012
Posts: 67,892
Likes: 3,831
From: Maryland
Lexus Fanatic
Joined: Oct 2003
Posts: 94,203
Likes: 218
From: Virginia/D.C. suburbs
Safe driving doesn't necessarily mean being attending a professional or high-performance driving school and/or being able to drive like Michael Shumacher. In general, it means being safe, obeying traffic-laws, not taking chances (especially with stupid chances), not using hand held phones or texting while driving, not driving DUI/DWI, and just using common sense. There is a reason why those who do this generally avoid chargeable accidents, and usually pay some of the lowest insurance rates in their class.
Last edited by mmarshall; Apr 11, 2026 at 07:23 PM.
Pomposity and self-aggrandizing aside, the reality is the bar for being a "better than average" American driver is low. General driving standards here are poor, the quality of driver training is low, and the driving tests are almost comically easy to pass. I learned to drive in another country where most people learn to drive in a manual and the test itself is much more exacting, and as a result the safety stats are better and the quality of driving - be it better lane discipline, not lane hogging, better general courtesy and awareness - is better.
FSD is a nascent technology, but I already consider it to have easily surpassed the driving ability of the average American. And while the low driving standards continue to decline, FSD and similar systems are continually improving.
FSD is a nascent technology, but I already consider it to have easily surpassed the driving ability of the average American. And while the low driving standards continue to decline, FSD and similar systems are continually improving.
You have it the wrong way, it is the opposite. The data would be skewed if you excluded those drivers, because those drivers are very much out there every day in the general driving population. See comments above about why this might be.
Of course there are going to be issues, FSD is a nascent technology. But again, you have the wrong basis for comparison. FSD has logged just under 9.2 billion miles at the time of posting. The statistics tell you that if you were to look at 9.2 billion miles of equivalent human driving, you are going to find more examples of human error than FSD error.
Tesla is the insurer for a large number of its own cars. If there were an accident, it may be the driver who is liable but is the insurer - often Tesla - who would ultimately be on the hook
It's not a question of dismissing anyone as a "hater" or anyone being "pro FSD people". The reality is, as noted several times above, this is still a nascent technology and things will go wrong. The argument is that things will go wrong fewer times with FSD at the wheel than they would with a person, "above-average" or otherwise. You only have to look at the enhanced FSD visualizations for a clue as to why. The system is capable of tracking a massive number of vehicles and other potential obstacles in every direction around you. It's tracking those behind and to the side of you, whereas people will tend to focus mostly on what's ahead. And again, FSD is getting better and better.
Bottom line - the argument isn't that FSD doesn't do stupid things. The argument is that it will do them less often than a person, and that while still nascent FSD is getting better at a rapid pace while the standard of human driving is gradually declining.
Bottom line - the argument isn't that FSD doesn't do stupid things. The argument is that it will do them less often than a person, and that while still nascent FSD is getting better at a rapid pace while the standard of human driving is gradually declining.
In addition, the statistics Tesla use includes every other non tesla car on the road made from 1930s - 2026. Many newer cars have a ton of safety features to prevent accidents. For example, blind spot monitor has a 23% reduction in accidents https://www.itskrs.its.dot.gov/2019-b01384.
So who is to say FSD is even better than a 2026 vehicle with all the safety features available and an average driver?
Out of the 10 billion miles of FSD, how much of it is during bad weather? Do FSD users commonly use it during snow / ice? torrential rain? How much percent of that 10 billion miles of FSD is during bad weather compared to the same 10 billion miles of humans? I would estimate 40-50% of FSD miles are in California, famous for great weather year round.
As I said, this statistics that is used is extremely favorable to Tesla. However, imo it has many holes which makes it inaccurate.
Lexus Test Driver
Joined: Mar 2013
Posts: 5,441
Likes: 946
From: IL
For someone to justify the claim of being an above average driver, they would need to have this quantified. Do you have a trophy? One of our daughters has taken driving skills training as part of her job in the government. One of them involved reversing at a high rate of speed and spinning the car around and slamming it into drive. She went through extensive days of classroom and on track training/testing. She loved it and came out thinking she was a better driver. I won't say this to her face, but she's not. She gets road rage easily, which doesn't help. She lives in Virginia, so be on the lookout for an "above average" driver flipping you off when you don't use your blinker to change lanes, or drive 10mph over the speed limit passed her.
Back to FSD. I don't know how many times those of us who have used it have to say it's not perfect. There's a reason it needs your attention, for those rare instances you need to intervene. I wonder how many of you naysayers would have seen the guy in the left lane in my blind spot drift half way into my lane. If FSD didn't move us over, he would have rammed us. Sure, you may have had your eyes on that one car out of many in your side mirror at the time, and known there was room on the right to swerve, but odds are you wouldn't 100% of the time like FSD will. One more time, FSD is a tool that helps with driver fatigue. It should not be used as a personal chauffeur.
EV ftw!!!
Joined: Nov 2002
Posts: 8,947
Likes: 545
From: Lake Ontario
Pomposity and self-aggrandizing aside, the reality is the bar for being a "better than average" American driver is low. General driving standards here are poor, the quality of driver training is low, and the driving tests are almost comically easy to pass. I learned to drive in another country where most people learn to drive in a manual and the test itself is much more exacting, and as a result the safety stats are better and the quality of driving - be it better lane discipline, not lane hogging, better general courtesy and awareness - is better.
FSD is a nascent technology, but I already consider it to have easily surpassed the driving ability of the average American. And while the low driving standards continue to decline, FSD and similar systems are continually improving.
You have it the wrong way, it is the opposite. The data would be skewed if you excluded those drivers, because those drivers are very much out there every day in the general driving population. See comments above about why this might be.
Of course there are going to be issues, FSD is a nascent technology. But again, you have the wrong basis for comparison. FSD has logged just under 9.2 billion miles at the time of posting. The statistics tell you that if you were to look at 9.2 billion miles of equivalent human driving, you are going to find more examples of human error than FSD error.
Tesla is the insurer for a large number of its own cars. If there were an accident, it may be the driver who is liable but is the insurer - often Tesla - who would ultimately be on the hook
It's not a question of dismissing anyone as a "hater" or anyone being "pro FSD people". The reality is, as noted several times above, this is still a nascent technology and things will go wrong. The argument is that things will go wrong fewer times with FSD at the wheel than they would with a person, "above-average" or otherwise. You only have to look at the enhanced FSD visualizations for a clue as to why. The system is capable of tracking a massive number of vehicles and other potential obstacles in every direction around you. It's tracking those behind and to the side of you, whereas people will tend to focus mostly on what's ahead. And again, FSD is getting better and better.
Bottom line - the argument isn't that FSD doesn't do stupid things. The argument is that it will do them less often than a person, and that while still nascent FSD is getting better at a rapid pace while the standard of human driving is gradually declining.
FSD is a nascent technology, but I already consider it to have easily surpassed the driving ability of the average American. And while the low driving standards continue to decline, FSD and similar systems are continually improving.
You have it the wrong way, it is the opposite. The data would be skewed if you excluded those drivers, because those drivers are very much out there every day in the general driving population. See comments above about why this might be.
Of course there are going to be issues, FSD is a nascent technology. But again, you have the wrong basis for comparison. FSD has logged just under 9.2 billion miles at the time of posting. The statistics tell you that if you were to look at 9.2 billion miles of equivalent human driving, you are going to find more examples of human error than FSD error.
Tesla is the insurer for a large number of its own cars. If there were an accident, it may be the driver who is liable but is the insurer - often Tesla - who would ultimately be on the hook
It's not a question of dismissing anyone as a "hater" or anyone being "pro FSD people". The reality is, as noted several times above, this is still a nascent technology and things will go wrong. The argument is that things will go wrong fewer times with FSD at the wheel than they would with a person, "above-average" or otherwise. You only have to look at the enhanced FSD visualizations for a clue as to why. The system is capable of tracking a massive number of vehicles and other potential obstacles in every direction around you. It's tracking those behind and to the side of you, whereas people will tend to focus mostly on what's ahead. And again, FSD is getting better and better.
Bottom line - the argument isn't that FSD doesn't do stupid things. The argument is that it will do them less often than a person, and that while still nascent FSD is getting better at a rapid pace while the standard of human driving is gradually declining.
Lexus Fanatic
Joined: May 2012
Posts: 67,892
Likes: 3,831
From: Maryland
Safe driving doesn't necessarily mean being attending a professional or high-performance driving school and/or being able to drive like Michael Shumacher. In general, it means being safe, obeying traffic-laws, not taking chances (especially with stupid chances), not using hand held phones or texting while driving, not driving DUI/DWI, and just using common sense. There is a reason why those who do this generally avoid chargeable accidents, and usually pay some of the lowest insurance rates in their class.
For someone to justify the claim of being an above average driver, they would need to have this quantified. Do you have a trophy? One of our daughters has taken driving skills training as part of her job in the government. One of them involved reversing at a high rate of speed and spinning the car around and slamming it into drive. She went through extensive days of classroom and on track training/testing. She loved it and came out thinking she was a better driver. I won't say this to her face, but she's not. She gets road rage easily, which doesn't help. She lives in Virginia, so be on the lookout for an "above average" driver flipping you off when you don't use your blinker to change lanes, or drive 10mph over the speed limit passed her.
I don’t have a trophy, I have certificates.
I just don’t see how someone who has taken the time to practice and be trained in how to drive when almost nobody else does that wouldn’t be considered an above average driver.
Thread Starter
Lexus Champion
Joined: Oct 2019
Posts: 19,428
Likes: 3,656
From: California
I've taken several golf lessons, but I can tell you that I'm far from an above average golfer. LOL
For someone to justify the claim of being an above average driver, they would need to have this quantified. Do you have a trophy? One of our daughters has taken driving skills training as part of her job in the government. One of them involved reversing at a high rate of speed and spinning the car around and slamming it into drive. She went through extensive days of classroom and on track training/testing. She loved it and came out thinking she was a better driver. I won't say this to her face, but she's not. She gets road rage easily, which doesn't help. She lives in Virginia, so be on the lookout for an "above average" driver flipping you off when you don't use your blinker to change lanes, or drive 10mph over the speed limit passed her.
Back to FSD. I don't know how many times those of us who have used it have to say it's not perfect. There's a reason it needs your attention, for those rare instances you need to intervene. I wonder how many of you naysayers would have seen the guy in the left lane in my blind spot drift half way into my lane. If FSD didn't move us over, he would have rammed us. Sure, you may have had your eyes on that one car out of many in your side mirror at the time, and known there was room on the right to swerve, but odds are you wouldn't 100% of the time like FSD will. One more time, FSD is a tool that helps with driver fatigue. It should not be used as a personal chauffeur.
For someone to justify the claim of being an above average driver, they would need to have this quantified. Do you have a trophy? One of our daughters has taken driving skills training as part of her job in the government. One of them involved reversing at a high rate of speed and spinning the car around and slamming it into drive. She went through extensive days of classroom and on track training/testing. She loved it and came out thinking she was a better driver. I won't say this to her face, but she's not. She gets road rage easily, which doesn't help. She lives in Virginia, so be on the lookout for an "above average" driver flipping you off when you don't use your blinker to change lanes, or drive 10mph over the speed limit passed her.
Back to FSD. I don't know how many times those of us who have used it have to say it's not perfect. There's a reason it needs your attention, for those rare instances you need to intervene. I wonder how many of you naysayers would have seen the guy in the left lane in my blind spot drift half way into my lane. If FSD didn't move us over, he would have rammed us. Sure, you may have had your eyes on that one car out of many in your side mirror at the time, and known there was room on the right to swerve, but odds are you wouldn't 100% of the time like FSD will. One more time, FSD is a tool that helps with driver fatigue. It should not be used as a personal chauffeur.