Only 1 crash for every 7.44 million miles driven using Tesla Autopilot in Q1 2025

[Gigahorse]

hard to know, sometimes I intervene just because I am too close to something and don't want to risk it but other times when I am driving alone I let it do it's thing and haven't hit anything as yet. Unsupervised is going to be cool....really looking forward to my first RoboTaxi ride.

I hope unsupervised is not another 10 years out for the CT, having been a FSD user for the last 4 years and it has improved, but not to the point I can take a nap.

Sponsored

 

[Michael S]

NEWS: Tesla has revealed that in Q1 2025, they recorded one crash for every 7.44 million miles driven in which drivers were using Autopilot technology. For drivers who were not using Autopilot technology, Tesla recorded one crash for every 1.51 million miles driven.

By comparison, the most recent data available from NHTSA and FHWA (from 2023) shows that in the US there was an automobile crash approximately every 702,000 miles.

Sources:

Tesla Vehicle Safety Report:
https://www.tesla.com/VehicleSafetyReport

I Western Washington, 17% of cars on the road are Teslas. Tesla accounts for the highest fatality rate per mile driven.
This is data from NHTSA

 

[jenkins007]

Meaningless unless one corrects for the bias that Autopilot gets used more often on roadways with lower accidents/mile; e.g. freeways vs. city streets. The only way to do this is to weight the Autopilot-off accident rate to represent the same mix of miles traveled on the same exact roadways, at the same time periods, as Autopilot-on miles.

I use mine in San Antonio city driving almost exclusively. Haven’t had a chance to use highway miles, except for maybe 5% of the time. And that’s about 8000 miles that are on my vehicle right now.

 

[riggster]

Meaningless unless one corrects for the bias that Autopilot gets used more often on roadways with lower accidents/mile; e.g. freeways vs. city streets. The only way to do this is to weight the Autopilot-off accident rate to represent the same mix of miles traveled on the same exact roadways, at the same time periods, as Autopilot-on miles.

Are you certain of that? I don’t use autopilot, but FSD instead. I use it a lot in the city. I don’t know the numbers of highway vs city miles autopilot is used. I’m not saying you’re wrong, just wondering where you got that information. What would account for the significantly lowered crash miles in Tesla vehicles where autopilot is not used? I would say that they could have easily provided a lot more information than they did, but i don’t think the data is completely useless, just severely lacking.

 

[Docedward]

I Western Washington, 17% of cars on the road are Teslas. Tesla accounts for the highest fatality rate per mile driven.
This is data from NHTSA

Yeah- that data is NOT from NHTSA- it was from a iSeeCars study (yeah, I’ve never heard of them either) covering 2018-2022. Just Groq or ChatGPT it. Bottom line for me- Tesla’s claims are a bit skewed to show safety- iSeeCars claims are way more skewed (proprietary estimation of mileage not open to public)- IMHO, it really depends 75% on driver skills and behavior and 25% on the safety of the vehicle.

 

[scottf200]

People have been questioning Tesla "method" of counting for a while.
You can search for this in Google/X/etc to see various questions: Tesla airbag NHTSA


https://static.nhtsa.gov/odi/inv/2022/INCR-EA22002-14496.pdf

 

[Beetlebug62]

I Western Washington, 17% of cars on the road are Teslas. Tesla accounts for the highest fatality rate per mile driven.
This is data from NHTSA

Okay, looked it up; and it's an iseecars analysis of NHTSA FARS data.

So, just to be clear, this is NOT Autopilot data, just general crash fatality data.

And this is the note they put at the bottom:

They conclude it's a driver behavior and conditions issue, not a design issue.

One last thing:

Seeing the urban data, implies to me that with further analysis, there's probably an urban-rural split. Urban areas are far more dangerous than rural ones. With the implication that if you are a manufacturer with lots of cars owned in urban areas, relative to other manufacturers, then your crash fatality data will skew higher.

They should normalize urban vs rural, to get a true comparison, in order to draw minimal conclusions regarding safety.

To create a sensational headline:

Is just clickbait, if the real story is that urban driving is more dangerous than rural driving, and that cars primarily driven in urban areas will have more fatal crashes.

 

[riggster]

I Western Washington, 17% of cars on the road are Teslas. Tesla accounts for the highest fatality rate per mile driven.
This is data from NHTSA

As I am writing this, I notice that others have already gone down this route and shown the data stated in this post.

The statement is false overall. The first part, claiming 17% of cars in Western Washington are Teslas, is not supported; estimates are around 1.5%.

The second part, claiming Tesla has the highest fatality rate per mile driven, is from the iSeeCars study using NHTSA data, but it's controversial due to methodological debates. The latest NHSTA data available is from 2023. During the iSeeCars study, they counted the number of fatalities using NHSTA data, but then used their "proprietary data" to estimate the number of miles driven. They have not been transparent about this part of their data for several years.

Anyone can download the NHSTA data here: NHTSA File Downloads | NHTSA

Tesla has supplied the data for the miles driven by their cars in 2023, and when you use tesla's reported miles driven, the data from the iSeeCars studies turns out to be bullshit. They're probably not transparent about their data because they did something to calculate the miles driven for each car like take the age of the car in years x 12000 to estimate the distance driven.

 

[riggster]

People have been questioning Tesla "method" of counting for a while.
You can search for this in Google/X/etc to see various questions: Tesla airbag NHTSA


https://static.nhtsa.gov/odi/inv/2022/INCR-EA22002-14496.pdf

NHSTA had the same question. Some time back they asked for data involving

 

[riggster]

You’re correct. In April 2021, NHTSA requested data from Tesla on crashes, including cases where Autopilot was disengaged before impact, as part of its investigation into Tesla’s driver assistance systems. The probe, expanded in August 2021 to focus on 11 crashes involving emergency vehicles, later noted 111 crashes where driver inputs inadvertently disengaged Autopilot. However, these investigations have not proven any systemic issues with Tesla’s Autopilot system.

Tesla’s Autopilot and FSD, already strong, continue to improve. Here’s a TechCrunch article discussing the investigation. While not Tesla-friendly, it does include factual data that supports Tesla's claims if you understand statistics well enough:

Tesla Autopilot investigation closed after feds find 13 fatal crashes related to misuse | TechCrunch

Statistics also support Tesla’s FSD and Autopilot, countering much of the misinformation out there.

Sponsored