Overnight, Waymo and Uber announced an expansion of their partnership. In a very surprising move, the service won’t actually be run by Waymo.
Waymo has had a long history of developing their Autonomous vehicles and has the service operating in Phoenix, Los Angeles, San Francisco.
Today’s announcement will see the locations you can order a Waymo driverless car, expand from 3 to 5 locations, adding Austin, TX and Atlanta, GA.
Today’s announcement is very different than previous expansions in that the Waymo service in these two new cities, will actually be run by Uber exclusively. Given the large potential revenue opportunities with a fully driverless robotaxi operations in a combined population of around 1.48 Million people, this seems like Waymo is giving up the golden goose here and it’s not clear why.
Waymo started at Google back in 2009 and was later spun out to its own company. Their technology approach has been to buy vehicles, most recently a Jaguar i-Pace and add lidar, radar, cameras and the compute necessary to operate the vehicle without a driver. The software, known as ‘Waymo driver’ requires a location to be scanned in detail to create HD maps. As the vehicles navigate the environment, they compare their current position to the map to help navigate safely and do have some ability to adapt to the changing world around them.
With many rides being completed successfully every day, perhaps the biggest challenge for Waymo is scaling the technology to more locations, along with making the economics of the business stack up.
The car they selected has a high starting price at around US$70,000 before their custom hardware and integration costs. There’s obviously network costs in running the system and ongoing development to improve their software.
While driverless cars should be far cheaper than a regular Uber, given the driver would normally be taking a significant percentage of the cost of a ride. The reality is, the costs associated the vehicles and the network (including scaling to new cities), means the cost to the rider remains high.
This is also made more complex by the fact the Jaguar I-Pace still retains it’s steering wheel and pedals, meaning the driver’s seat is not an available seat for passengers. With less passengers per ride, the average cost of a ride remains higher.
This suggests that Waymo may be looking to distribute some of those costs to Uber, so why would Uber want to take this on?
Uber are walking a very delicate line here, by moving heavily into the autonomous vehicle space, this will be seen as competing with drivers who operate their own vehicles on the Uber network in these cities. Given the feedback I’ve seen online already, this is not going down well with drivers.
Unfortunately Uber lacks the ability to develop this technology themselves, so if they want to derive more of the revenue from each ride, using Autonomous vehicles to deliver passenger rides, will mean ditching their human drivers, they simply won’t be able to compete on price in the long-term.
If Uber do nothing, they will see their market share be taken by autonomous vehicle companies like Waymo, Cruise and others, so perhaps, at least for now, Waymo and Uber need each other, but I suspect in the long term, this relationship will be one plagued by competitive tensions and may lead to an acquisition or bankruptcy.
The Bloomberg article shared by Conor confirms the exclusivity of the Waymo Austin and Atlanta expansions which will be run by Uber.