While this is reported to have been a case where the driver had the car in drive instead of reverse, and somehow panicked and pressed the accelerator all the way to the floor, the question still needs to be asked: how do we know when a car is out of control NOT because of driver inputs, but rather my mechanical or software flaws in the vehicle itself? Lawsuits are being fought right now, and NHTSA and other federal agencies are hard at work studying the problem. But for the people trying to understand what happened from media reports, this is a limitation that needs further study.
The Storefront Safety Council is starting a research project on this topic of vehicle failure versus driver error. We are doing so for a variety of reasons, the most basic of which is that it is important to be accurate about causation in order to be accurate about prevention. In terms of preventing injuries to people on sidewalks or inside stores, the causation of a vehicle-into-building crash is less important than the means of protecting people and property against errant vehicles in the first place. Bollards or safety barriers do not care if the vehicle has jumped the curb due to being in the wrong gear, or the driver under the influence, or a mechanical or other failure with the car -- the bollards just stops the crash from happening. But in terms of understanding what is going on and for policy makers to make best use of information, that information and underlying data needs to be accurate.
Recent high profile accidents that have had these types of issues are many -- our resources are limited but we will do our best to find out what we can and report the results here and on the Storefront Safety Council website.