Autonomous Vehicles or Zombie Cars?: Towards a Philosophy of Transportation

"Such 'programming codes' furnish the vehicle with a feigned autonomy. In truth it’s not a vehicle anymore but something ghostly, a technological zombie that simulates knowledge and responsibility. Such zombies can be hacked and used as killing machines." ~ Rafael Capurro.

In Autonomous Zombies are not an Option, a recent interview with German fleet managing journal Flottenmanagement on the ethics of autonomous vehicles, Professor Rafael Capurro correctly places responsibility for the actions of driverless cars. By realigning unrealistic expectations of an "ethics" for autonomous vehicles, Capurro reminds us that autonomy encompasses more than a mere set of programmed rules. The admonition, as such, should weaken the focus of our inquiry into AI and remind us that there are other concerns in the consideration of autonomous vehicles beyond algorithmic decision making. The truth is, the conversation around autonomous vehicles should comprise one piece of a much larger consideration of a philosophy of transportation.

Having followed the conversation on autonomous vehicles for a number of years now, seeing the issues and concerns mount exponentially, I have begun to sense more and more the impossibility of creating a functional system of infrastructure and autonomous vehicle any time soon, at least in terms of fully autonomous vehicle systems/transportation systems (the two must be considered together). There are unprecedented factors to consider in the task, factors within factors, assuming that all factors can even be predicted (which, as Capurro points out in his article, they can't).

Capurro's article really brings home the complexity of the endeavour of autonomous vehicles by highlighting even just a few scenarios. His vision of "something ghostly, a technological zombie that simulates knowledge and responsibility", is disconcerting on a number of fronts, especially when we consider, as he notes, that "nobody is able to predict what mobility will look like in twenty, fifty years, let alone a hundred years".

Any transformation of the transportation system will be, by necessity, a slow transition over many decades. I think as engineers, technologists, policymakers and philosophers, we do ourselves a disservice by focusing too much on the idea of "ethics" for vehicles, rather than on considerations of infrastructure and culture that must play into such a transformation. Of course we must consider the simulation of ethics for AI, but not only so; and perhaps more pressing than an ethics for AI, we need to consider our ethics (human ethics!) first and foremost, since, after more than 2000 years, we are still working out among ourselves what it means to be with others. Are we sure we're ready to pass the torch of ethics to artificially intelligent systems? 

As Capurro notes, these supposed "ethics" are really only rules we place into systems. We are far from creating a type of autonomy for vehicles that enables a true ethics for decision making. I think instead we need to consider first several other factors in developing a philosophy of transportation, factors that take into account how best to rebuild society around new appropriate transportation models (models that take into account globalization, urbanization, overpopulation, trans-border commutes, etc.)

It occurs to me that we have chosen to base our ideas of developing autonomous vehicles on outdated transportation models that no longer work in global and overpopulated societies. Most urban planning and city development now recognizes the futility and non-feasibility of continuing with transportation models that are over 100 years old; developers are actively moving to reduce or eliminate the use of personal vehicles from city centres by restructuring road and walkway systems.

In other words, there is a perfect transportation storm growing and it has nothing to do with whether vehicles are autonomous or not. I think we need to address a philosophy of transportation holistically. This may mean shelving the autonomous vehicle discussion for now (for the time being) to focus on a top down approach to transportation that includes far more than the consideration of autonomous technologies. 

The problem is, nobody is about to shelve the autonomous vehicle project anytime soon so long as technology advances and profit awaits. For the same reason that IoT applications are in full speed production, despite the ever-increasing occurrence of malware attacks and hacking, so too will the autonomous vehicle find its way into a world ill-prepared to receive it. As such, the ethics work that Rafael and others are doing in the area of autonomous vehicles is critical and more important than ever. Perhaps it is from the platform of developing an ethics for autonomous vehicles that philosophers, technologists and policymakers can also advocate for a holistic philosophy of transportation.