The NHTSA has closed its investigation into the Tesla crash in Florida that killed Joshua Brown, declaring that it found no defects in the driver-assist software, and declining to issue a recall. Calling the regulator out on its “unseemly haste,” the Consumer Watchdog has taken a very aggressive stance on the ruling, which smacks of political motivation.
In a review of many other Tesla crashes, the NHTSA failed to identify “any incidents in which the systems did not perform as designed,” but noted that this transitional phase to autonomous driving still requires the “continual and full attention of a driver.” The Consumer Watchdog, meanwhile, sounds like it is clamouring for a scalp – demanding that Tesla CEO Elon Musk be held responsible.
Given the 74mph cruise control (in a 65mph zone) that was set by the driver, the investigation found that the driver should have had at least 7-seconds to react to the truck crossing the freeway. A Harry Potter movie was playing in the car, and the driver, a former Navy Seal, was the only occupant. The investigation found that the car made no attempt to avoid the accident.
It’s hard not to look at the politics of this, when it comes to the Consumer Watchdog’s statements. We’ve always taken the stance that Tesla has been pretty clear about reading the instructions and keeping your eyes on the road, but we’ve also criticised its choice of name. Autopilot is definitely somewhat misleading, when it comes to connotations, synonyms, and implications.
“NHTSA has wrongly accepted Tesla’s line and blamed the human, rather than the ‘Autopilot’ technology and Tesla’s aggressive marketing,” said John Simpson, the Consumer Watchdog’s Privacy Project Director. “The very name ‘Autopilot’ creates the impression that a Tesla can drive itself. It can’t. Some people who apparently believed Tesla’s hype got killed. Tesla CEO Elon Musk should have been held accountable.
We think it’s pretty obvious that most junior lawyers would be able to get Musk off the hook if that argument ever made it to court, given that the video evidence strongly suggests that the driver was not paying attention to the road at all. In the wake of the crash, Tesla did back off of the Autopilot branding, rolling back its prominence, but the company has always played by the rules when it comes to user guidelines and insisting that drivers maintain their focus on the road at all times.
Tesla has also updated the system to rely more on radar than image-recognition, and had a public falling-out with Mobileye, the provider of the camera technology used by the cars. It added a strike-out system that would disable the feature if it thought the driver was not paying attention.
Now, Tesla firing back with a libel/slander threat would prove entertaining, but we are pretty certain that the Consumer Watchdog doesn’t have a leg to stand on that wouldn’t be completely knee-capped by a competent lawyer. The NHTSA also notes that Tesla fully cooperated with its investigation.
The Consumer Watchdog’s Simpson accuses the NHTSA of being “simply too cozy with the industry it is supposed to be regulating.” Given the politicking that will be going on with the changing of the guard in the White House, this aggressive stance could be simple posturing, but we will be keeping tabs on whether the Consumer Watchdog does anything more than weep, wail, and gnash its teeth.
So the question becomes whether you trust the NHTSA to properly investigate such matters. This is the same body that effectively forced Comma out of the market, simply by asking it to provide some detail and documentation for its retrofit self-driving tech, and also found that once Tesla had installed Autopilot’s auto-steering tech in its vehicles, it saw a 40% reduction in crashes – going from 1.3 crashes per million miles to 0.8, according to the NHTSA’s Office of Defects Investigation report.
The regulator says it plans to aggressively oversee the introduction of new self-driving technologies, and has noted that its finding doesn’t constitute a get-out-of-jail-free card when it comes to autonomous systems that do cause crashes when they function incorrectly – meaning they wouldn’t be able to dump all the liability on the human driver.
But that 40% reduction in crashes is a pretty solid endorsement of the Autopilot tech. While directly comparable figures are hard to come by, New York’s DoT lists mainline and juncture accidents at around 2.81 accidents per million-vehicle-miles driven, and the NHTSA has listed the rate of accidents with property damage only (that is, no injuries) at 0.38 per 100,000 miles driven. That translates to 3.8 per million vehicle-miles driven, and there’s the caveat that around half of all such crashes aren’t reported to consider.
As for deaths, the US rate is around 1.12 fatalities per 100 million miles driven, which translates to 35,092 deaths in 2015, with the estimated number of people injured on the roads at 2.44m. As the NHTSA points out, in 2006, some 42,708 people were killed on the roads, and the current level of deaths per million-miles-driven has trended downwards from its height of 5.3 in 1965.
So for Tesla, that single death in the US really skews its fatalities per million-vehicle-miles-driven, but that 0.8 crashes per million-miles-driven isn’t directly comparable with the property-damage-only figures from the NHTSA – as presumably, there were injuries in crashes that triggered the car’s air bags, which was the criteria for the NHTSA’s analysis. However, a 40% reduction in crashes is certainly not to be sniffed at.