Uber – Fatal disengagement.

Reply to this post

Google is 5,000x better than Uber at autonomous driving.

• Although Google is suing Uber for the alleged theft of its Lidar (key autonomy sensor) design, it does not seem to have helped Uber much as it appears to be by far the worst at autonomous driving.
• This is still the case when one includes the regular car companies that most people have written off as having very little to offer in the new world of digital and autonomous cars.
• The best measure of an autonomous driving solution is how often the driver has to take over to correct shortcomings in the autonomous driving software.
• Regulations in California require those that test in the state to submit this data but typically, they all submit it in different ways (see here).
• There are also different types of disengagement such as when the car is going to hit something (critical) or when the safety driver feels uncomfortable (ordinary).
• Furthermore, companies test their cars in different conditions meaning that the data can really only be used as indication.
• However, the contrasts are so stark that I think that meaningful conclusions can be drawn about how advanced the autonomous driving solutions from different players really are.
• In order of performance the data shows:
  • No. 1 Waymo (Google) is 8x better than the number 2 with 1 disengagement every 5,128 miles driven.
  • Waymo has also driven at least 155x more miles (635,868) in the last 12 months than anyone else, meaning that it has collected more training data than all the others put together.
  • No. 2 BMW with 1 disengagement every 638 miles driven (8x worse than Waymo) but it only drove 638 miles raising questions to the validity of this data.
  • No. 3 Nissan with 1 disengagement every 146 miles and a total of 4,099 miles driven.
  • No. 4 Tesla with 1 disengagement every 3 miles with 550 miles driven but almost all of these occurred in wet road conditions.
  • I think that Tesla deliberately went out to push its system to the limit as wet conditions are known to be far more difficult for autonomous systems.
  • Hence, I do not think that is necessarily an indication of Tesla’s real position in the pecking order.
  • No. 5 Mercedes with 1 disengagement every 2 miles driven with 673 miles driven in total.
  • No. 6 Uber with 1 disengagement every 1 mile driven with a total of 20,354 miles in total.
  • Uber has just suspended its autonomous testing following a serious crash in Arizona despite the fact that it appears that the Uber vehicle was not responsible for the incident.
  • Uber has also been banned from testing in California for failing to register with the DMV.
• This is yet another indication that the key to artificial intelligence (the heart of all autonomous driving systems) currently is the amount of time that one has been working on the algorithms as well as the amount of data collected (see here).
• I am certain that this is why Waymo is the best because it began working on autonomous driving in 2009 (far earlier than anyone else) and in the last 2 years has driven more than 150x more miles than anyone else.
• The fact that Uber, by quite some margin, ranks last is potentially a serious problem in the long-term.
• This is because if the car companies have their own self driving technology (or use Google) then Uber may find itself being a middleman that is no longer required.
• Uber currently has the advantage because it has already established itself as the market place for drivers and passengers to transact and these types of positions are extremely hard to disrupt once created.
• This is why Uber commands the $70bn valuation that it does but unless it gets a handle on autonomous driving, this market place may become obsolete when humans stop driving cars.
• I still think that the technology will become mature long before the market is ready to adopt it (see here) meaning that Uber should be able to pick-up a viable solution at an attractive price once 2020 deadlines are missed and funding runs out.
• Despite this view, this is a key risk for Uber and one I would be uncomfortable with especially if I had put some money into the company at $62.5bn.