Tesla has began to release its quarter Autopilot safety reports after stopping for a 12 months. The automaker claims some improvements.
Since 2018, Tesla has been attempting to create a benchmark for its improvement in Autopilot safety by releasing a quarterly report that compares the variety of miles per accident on Autopilot versus off Autopilot.
The information was all the time limited and criticized for not considering that accidents are more common on city roads and undivided roads than on the highways, where Autopilot is usually getting used.
Nonetheless, it was still helpful to match it against itself over time and see if there have been any improvements, and there have been some incremental improvements at times.
Tesla suddenly stopped releasing those quarterly reports in 2022 with none explanation.
Now, the automaker has decided to start out it up again and released reports for all quarters as much as Q3 2022.
Tesla wrote for essentially the most recent data:
Within the third quarter, we recorded one crash for each 6.26 million miles driven during which drivers were using Autopilot technology. For drivers who weren’t using Autopilot technology, we recorded one crash for each 1.71 million miles driven. By comparison, essentially the most recent data available from NHTSA and FHWA (from 2021) shows that in america there was an automobile crash roughly every 652,000 miles.
That’s in comparison with one crash for each 4.35 million miles driven with Autopilot technology back in Q4 2021 – the last time Tesla was frequently releasing the info.
Tesla was also kind enough to plot the info right into a chart this time for higher visualization:
As you possibly can see, it goes up and down, but that’s partly seasonal. There are notoriously more accidents throughout the winter resulting from road conditions and since it gets dark sooner.
Since Tesla stopped gathering the info, the automaker has also significantly grown its Full Self-Driving Beta program, which actually allows the usage of more Autopilot technology on city roads.
Nonetheless, it’s unclear whether Tesla includes this data on this report.
Electrek’s Take
I do know. It’s a really limited dataset, and I too wish Tesla could be more transparent. Nevertheless it’s the very best we’ve right away, and it does show some improvements.
That’s what we’ve to work with for now.
As I recently reported, I genuinely hope Tesla would release more data specifically about its FSD Beta program so we will start seeing some solid numbers constructing a path to Tesla delivering on its self-driving promise.
The corporate restarting to release these reports is likely to be a step in that direction. We are going to see.