See those spikes? That's when crashes happened.

The first spike was June 26 - July 2, 2016.

Tesla car mangled in fatal crash was on Autopilot and speeding, NTSB says
The National Transportation Safety Board has released a preliminary report and photo concerning the first known fatality of a Tesla car using Autopilot.

The second spike was March 25 -31, 2018.

Tesla was on Autopilot in fatal crash
The company says a Model X vehicle involved in a fatal crash in the US was in Autopilot mode.

These two incidents were the worst of 4 fatalities[1,2], as of October 2019. People think that Tesla Autopilot is dangerous because of the extensive media coverage, but it's not true.

There were 34,247 fatal motor vehicle crashes in the United States in 2017 in which 37,133 deaths occurred. This resulted in... 1.16 deaths per 100 million miles traveled. - IIHS, December 2018

As of October 11, 2019, ~1.88 billion miles have been driven with Autopilot on.[3]


Let's do the math. 4 deaths for 1.88 billion miles. Autopilot was released in October 2015. So 4 deaths divided by 4 years. 1 death a year, on average. Now, haw many miles were estimated to be travelled with Autopilot in 2018? 573,827,165.50, or ~573 million. That statistic can be calculated from Fridman, who has a CSV of his estimations on his website, as well as the chart above.

1 death per ~573 million miles travelled on Autopilot. 1.16 deaths per 100 million miles traveled in all vehicles in the U.S. or...

~0.17 deaths per 100 million miles travelled on Autopilot compared to 1.16 deaths per 100 million miles traveled in all vehicles in the U.S. in 2018. Statistically, Autopilot is much safer, but there are other factors at play. For example, chances are that Autopilot is used more on highways than in city streets, because it works much better on highways. That being said, more accidents happen on rural roads than on urban roads, or in other words, more accidents happen on quiet streets than on busy streets.[4]

Although 19 percent of people in the U.S. live in rural areas and 30 percent of the vehicle miles traveled occur in rural areas, almost half of crash deaths occur there. - IIHS, Fatality Facts 2017

Long story short, don't jump to conclusions. Tesla's Autopilot system isn't perfect, but it is safer than human drivers. Don't just take my word for it, CNET has an article on it too. I encourage you to read it. The author, Ashley Esqueda, owns a Tesla Model 3. Here's some snippets from that article.

One of [my family members] actually said, "turn that death trap in immediately and get a Volvo, or something else SAFE."
No, Tesla’s Autopilot isn’t dangerous, you just have to use it correctly
Commentary: A lot of folks who don’t own Teslas say Autopilot is unsafe. But I do own one, and used correctly, it’s perfectly fine.

Questions, concerns, or comments? Message me on Instagram at @amitchell516


Image: Bram Van Oost from Unsplash

1. Wikipedia. List of self-driving car fatalities. Accessed November 4, 2019.

2. Tesla Deaths. Every Tesla Accident Resulting in Death. Accessed November 4, 2019.

3. Lex Fridman. Tesla Vehicle Deliveries and Autopilot Mileage Statistics. Accessed November 4, 2019.

4. Insurance Institute for Highway Safety (IIHS). Urban/rural comparison. Accessed November 4, 2019.