In several places, this statement is missing some important information, which leads to the impression that the story is being spun to imply the driver had more

In several places, this statement is missing some important information, which leads to the impression that the story is being spun to imply the driver had more responsibility in this crash than may have been the case. The statement says "The driver had received several visual and one audible hands-on warning earlier in the drive and the driver’s hands were not detected on the wheel for six seconds prior to the collision. The driver had about five seconds and 150 meters of unobstructed view of the concrete divider with the crushed crash attenuator, but the vehicle logs show that no action was taken." (my emphasis.) What is not stated here is how long before the impact was it clear that the car was off course and was not going to make a correction - i.e. how much time would an attentive driver have to realize that something was going wrong? Tesla may not yet know the answer to that question, but if it does not, I do not think it should suggest that the driver had five seconds to respond (the only information I have read on this matter is Tesla's own statements, so I do not know if other reporting on the crash shows that this was the case.) Hands off the wheel does not prove that the driver is inattentive. More importantly, hands on the wheel does not show that the driver is attentive. IMHO, something at least as effective as eye tracking is necessary to establish that. Furthermore, Tesla's own statement indicates that it is allowing too long a period of inattentiveness, even for highway driving. It is reasonable for Tesla to use statistics to justify the utility of their systems, so long as they compare like with like. Unfortunately, this statement is cagey about that: "If you are driving a Tesla equipped with Autopilot hardware, you are 3.7 times less likely to be involved in a fatal accident." Compared to what? Teslas without such hardware? Regardless of whether the hardware is being used? depending on what they are comparing it to, these statistics might be on account of simpler, more reliable safety features of the vehicle, or even owner demographics. Finally, the absence of a working crash attenuator, while a tragic fact, has no bearing on the issue of Autopilot's reliability.