Opinion | Autonomous transportation at scale is here4 min read . Updated: 14 Aug 2018, 09:45 PM IST
Just like ride-hailing services do today, it is inevitable that autonomous transport will primarily cater to affluent sections of society
Google’s self-driving project, Waymo, began in 2009. Since then, the fleet has covered more than 5 million miles. According to the Waymo blog, it took them six years to cover the first million miles but just over a year to cover the next million. The third took them under eight months and the fourth just six. The last million miles were completed in less than three months. Today, the entire fleet covers over 10,000 miles per day.
In March, Waymo announced it will buy 20,000 electric self-driving vehicles from Jaguar. If you add that number to the sum of all the other cars they have committed to deploy, it is clear that we are about to witness a rapid increase in the number of autonomous vehicles on our roads.
It is estimated that even though no more than 20,000 people on the planet have, to date, been driven by a robot-chauffeur, in three years time, over a million people will take a ride in a driverless car every day.
Over the last couple of years, I have devoted space in this column to talking about the urgent need to regulate autonomous vehicles. I have written about why we need to establish the ethical basis on which autonomous vehicles should be designed—particularly in connection with the choices that need to be made around protecting passengers at the cost of harming pedestrians. I have suggested that we evolve a regulatory framework that companies should be compelled to abide by while they make these programmatic choices rather than leaving it up to them to determine at their own discretion.
All this while, I had assumed that we were still a long way away from that tipping point when autonomous transportation became the norm rather than the exception. I had assumed that, unlike most legislative exercises where we are forced to make laws to address societal problems that are already upon us, with self-driving cars we would at least have the time to properly consider the first and second order impact of the technology before enacting a new legislation.
That timeline now seems to have been overly optimistic. It is becoming increasingly apparent that the last person to be issued a driver’s licence has already been born.
What this means is that it is probably already too late to regulate the basis on which manufacturers approach the hard ethical problems surrounding autonomous transportation. Those choices have already been decided by vehicle manufacturers who have begun to put cars on the roads on that basis.
While they may not be able to guarantee that these vehicles will never get into an accident, manufacturers seem to be willing to take that risk on the basis that, in general, vehicles are far safer when under the control of autonomous systems than human drivers. They are probably also betting on the fact that if we can get more autonomous vehicles on the roads fast enough, we will quickly arrive at the point where they outnumber human-controlled cars—at which point, traffic will become much more predictable.
This means we have already entered that messy transitional phase in which autonomous vehicles will coexist on our roads with manually controlled vehicles. Any changes we make to our laws to deal with this period are going to be temporary at best—and only relevant until autonomous vehicles have scaled up to be the dominant form of traffic on our roads.
What we can, and probably should, do is think about how this will alter society as we know it. Transport is such an important part of our urban lives that the impact of replacing drivers en masse with robots will not be inconsiderable. And yet, as this technology becomes mainstream, cities are likely to actively discourage human drivers once the economic efficiency and safety of all-autonomous city transport becomes evident.
The consequence of these decisions on the urban labour market will be harsh but inevitable.
None of this is likely to happen all at once. The current design of our roads, with its traffic lights and parking areas, caters to cities filled with human drivers. It will take time to adapt these spaces to address the requirements of safe, autonomous transportation. As a result, our cities will be slowly remade to better cater for autonomous vehicles that will carry us from our doorstep directly to wherever we might want to go.
We will see the creation of dedicated vehicle lanes physically segregated from pedestrian traffic (much like our urban commuter railway lines are today) to limit accidents completely. We will not mind that our urban spaces are carved up in this manner—so long as we can be picked up from our doorstep and taken exactly where we need to go.
Just like ride-hailing services do today, it is inevitable that autonomous transport will primarily cater to affluent sections of society. It is, therefore, likely that urban redevelopment will first take place in those areas of the city where people who can best afford this new form of transportation reside and work. This will result in increasingly segregated cities where the rich will use efficient autonomous transportation while poorer inhabitants are forced to commute using an increasingly underfunded public transport system.
If you couple this with all the rapid unemployment that will inevitably result from replacing human drivers with robots, we could be hurtling towards the very dystopia that science fiction has been warning us about.
Rahul Matthan is a partner at Trilegal and author of Privacy 3.0: Unlocking Our Data Drive Future. Ex Machina is a column on technology, law and everything in between. His Twitter handle is @matthan.
Comments are welcome at firstname.lastname@example.org