While many says it's the ultimate meme stock, I also can't help thinking it's the ultimate manipulation stock - it seems some people desperately want it to go down while others desperately want it to go up.
While the linked article is playing up the tesla angle (and so may be thought to be manipulative) the underlying study does not seem to be unusually focused on tesla, it's simply listing the results of a fairly straightforward analysis. I also have no reason to doubt it as I more or less expected tesla to have bad fatality rates compared to class (although I guess I wouldn't have expected them to be quite this bad - I thought they'd be bad compared to other luxury vehicles of similar weight size and price, not absolutely bad compared to most cars).
But you can find the underlying numbers and critique them if you have reason to think they might be wrong. E.g. if you believed the claims that autopilot was safer than human drivers and was saving lives, you might have expected to see a sign of that in this data (I didn't).