{{ reaction times using screens while driving are worse than being drunk or high }}
Latest Tesla safety report puts FSD as 10 times safer than a human driver.
intercst
5 Likes
Driving around town here makes me want to see FSD become a reality. Distracted drivers are everywhere. Baby boomers ( and Iâm one of them ) need to just concentrate on the task at hand, which should be driving. But they seem to be doing everything but.
2 Likes
I hate touch screens! Very distracting. My Impreza has buttons for all the major controls. I know exactly where they are and donât even have to look.
Wendy
Someday, in a history class, a student will say, âWere they crazy to let humans drive cars when there were 30,000 fatalities every year?â
Wendy
4 Likes
I sometimes wonder that myself right now.
And what will that history student say when the professor shows them that it used to be 50,000 fatalities every year? With fewer total miles driven by a population that was 2/3 of the population today (using 1980 numbers)?
âPeter
1 Like
Heck, when I moved to Houston in 1981, they had drive through Package Stores, and I donât believe it was illegal at the time to open up a brewski for the drive home as long as you âblewâ below the legal limit for a DUI.
intercst
3 Likes
Then they cracked down and made it so open containers were only legal in the back seat.
2 Likes
No doubt, a totally unbiased safety report.
8 Likes
Thatâs actually pretty good data and is consistent with the safety improvement of other manufacturersâ self-driving systems.
{{ Waymoâs report centered on incidents between its vehicles and âvulnerable road users.â Compared to human drivers, Waymoâs self-driving, electric Jaguars encountered 92 percent fewer crashes with pedestrians that resulted in injuries, 82 percent fewer crashes with cyclists involving injuries, and 82 percent fewer crashes with injuries that involved motorcyclists. There was also good news for car-to-car crashes, with 96 percent fewer injury-involving intersection crashes among Waymos, compared to human drivers, and 85 percent fewer crashes with suspected serious or worse injuries. }}
Waymo is also opening a plant in Arizona to manufacture Robotaxis, including for sale to the public. (l bet theyâll be expensive.)
intercst
The data do not say anything (rigorous) about FSD per se.
The fairest comparison from the data (most apples to apples), is Tesla driver
with Autopilot (1 accident per 7.44 million miles)
and
without Autopilot (1 accident per 1.51 million miles)
(as this somewhat controls for vehicle form and driver âdemographicâ - any attribute of driver relevant to performance, such as age, accident history, education, income, etc). Although the data are still correlative (eg, there could be something different about Tesla drivers who choose or choose not to use Autopilot).
What the data do not tell us:
- performance of Tesla FSD (no human intervention) per se
- performance of Tesla Autopilot versus any other driver assist texhnology
I am assuming Autopilot is not the same as FSD with no human intervention.
4 Likes
Yes. I should have wrote âAutopilotâ rather than FSD.
Latest 1Q2025 report puts Autopilot at 10X safer than the US average driver.
https://www.tesla.com/VehicleSafetyReport
intercst
1 Like
Then, when they cracked down on that, people with hatchbacks would pull down the back seat and put it in the âtrunk.â
2 Likes
I would re-phrase that as
âLatest 1Q2025 report puts average Tesla driver using Autopilot is 10x safer than the average US driver.â
At least 3 factors differ between the two things we are comparing:
- Tesla vehicle vs âaverage US vehicleâ
- Tesla driver vs âaverage US driverâ
- Tesla driver assist tech (Autopilot) vs âaverage US driver assist tech (which includes case of no driver assist)â
In ideal world, like clinical trials, we change 1 factor (eg, FSD) and make the measurement (accident rate) with and without that 1 factor (along with other good experimental practice).
In reality for AVs, weâll see what the regulators adopt.
2 Likes
Fair enough.
But those are sample sizes in the millions. Itâs hard for me to believe that the average Tesla driver (not using Autopilot) is more or less safe than the population at large
intercst
1 Like
Sample size is only one part of experimental design.
With a large sample size, but a lack of other aspects of good statistical design, all youâve accomplished is many uninformed measurements.
Maybe. Ask the auto insurers.
Certainly the average Tesla driver deviates demographically from the average driver (I would expect). Whether these demographic differences also translate to differences in accident rate (all else equal, such as vehicle), is another question.
Iâm open to whatever the data show.
3 Likes
Dear Wendy
One of my sisters bought the Hyundai EV almost 2 years ago for the buttons.
Theirs and Tesla rank the best.
Interesting point. Seems like thereâs some fuzziness that isnât completely straight forward.
Shifting topics a bit - Now, insurance for a Tesla is often significantly higher due to repair costs and complexities related to technology. As AVs become more widely available, wouldnât manufacturers have to pick up some, if not most, of the insurance cost?
Why should owners pay higher premiums for an AV if Iâm theyâre not driving? Why should owners pay premiums for cybersecurity risk that can lead to their AV going on a Highlander type road rage killing spree? Is there a potential for AV manufacturers to get into the insurance biz?
I donât carry collision & comp on my vehicles since I can afford to replace it out of unspent annual dividend income. All I have is liability coverage ($1 MM limit)
When I moved up from a 250 HP V-6 Nissan Altima to a 428 HP Model Y, State Farm didnât increase my insurance premium. (Presumably, in the wrong hands 4.2 sec 0-60 acceleration could do more damage to the public.)
intercst
I think it goes without saying, youâre exceptional.
1 Like