IEEE Technology and Society Magazine - June 2016 - 42

with General Motors whereby GM is developing self-driving vehicles for Lyft's on-demand transportation service
and funding an additional $500 million to help Lyft keep
up with Uber's relentless expansion [12].

Social Inflection and Ethical Concerns
Dr. Jeffery Greenblatt of the U.S. Department of Energy's
Berkeley Lab estimates the widespread adoption of automated vehicles into the U.S. taxi fleet will reduce greenhouse gas emissions by 87-94% on a per vehicle basis by
2030 as compared to today's taxis [13], [14]. Greenblatt

Companies like Uber are betting that
private individual ownership of cars
will wane with the emergence of
driverless vehicles.

bases his estimate on the observation that 87% of U.S.
light duty vehicle trips transport only 1 or 2 people.
Therefore, elimination of the driver in automated taxis
will result in taxi fleets dominated by small, two-occupant
vehicles. Additionally, the driving habits of these smaller
automated vehicles will be tuned for economy, and their
diminutive size will be well suited for electric battery rather than fossil fuel power. Finally, the low cost per mile to
ride in small, electric, driverless cars will encourage widespread use of ubiquitous on-demand fleets of self-driving
taxis in place of privately owned vehicles.
Among industry followers, Greenblatt is not unusual
in his estimate that private individual ownership will
wane with the emergence of driverless vehicles. Companies like Uber are betting on it, and - given that the
average American household spends 19% of their
income on transportation [15] - the economics will
press in that direction. Aside from economic rationality,
the availability of ubiquitous, affordable, on-demand,
automated personal transportation is likely to lure some
folks away from car ownership by simplifying their lives.
They will no longer have to deal with parking, fueling,
maintenance, registrations, breakdowns, asset depreciation, road stress, and traffic citations. Additionally, the
portion of their day previously spent driving will become
available for leisure or productive work while in transit.
At the recent Automotive Tech.AD 2015 conference
devoted exclusively to vehicle automation, attendees
seemed infused with excitement at the prospect of participating in what they see as an imminent social

inflection point [16]. If things turn out as expected, the
combination of waning personal vehicle ownership,
increased leisure or productive time, reductions in
greenhouse gas emissions, widespread reductions in
traffic deaths, and pervasive trust in transportation
"robots" will amount to nothing less than a paradigm
change in the Western lifestyle.
Yet - serious work has barely begun on the ethical
issues associated with the driving decisions that autonomous vehicles will be required to make. In essence,
vehicle manufacturers will be shouldering the moral
responsibility and legal liability for preprogrammed driving decisions that, in the past, have been the responsibility of human drivers. The technology is getting ahead
of our social and legal frameworks.
California Polytechnic State University philosopher
Patrick Lin points out that our laws are ill-equipped to
deal with autonomous vehicles. He illustrates the difficulty by exploring variations of Philippa Foot's famous
trolley problem [17]. Imagine a runaway trolley hurtles
toward five people on its track. You have the opportunity to prevent the certain death of those five by actuating
a track switch to change the trolley's course; however, in
doing so you would be sending the trolley onto another
rail occupied by one person. Is the correct ethical
choice to stand by and watch the five die or to actuate
the switch causing the one to die? Autonomous vehicle
software will necessarily make similar - usually more
subtle - ethical choices.
If, for the sake of illustration, we suppose that the
sacrifice of the one to save five is, in effect, "using" him
against his will, the trolley problem can serve as a contrast between utilitarian and libertarian ethics. Actuating the switch to save the five is an ethical stand
reflecting John Stuart Mill's utilitarian view that our
actions are right in proportion to the happiness they
promote and wrong in proportion to the unhappiness
they promote [18]. Under utility, our decisions should be
guided so as to bring about the greatest good for the
greatest number of people, and that guidance is a core
tenet of modern liberalism. In contrast, a refusal to sacrifice the one to save five would be an ethical stand
reflecting Immanuel Kant's principle that individuals are
ends in themselves and not means to the ends of others
[19]. Libertarian philosopher Robert Nozick further
developed Kant's principle by insisting that, if an
infringement on another person for one's own ends is
wrong between individuals, it is just as wrong between
society and individuals. To Nozick, respect for the rights
of individuals is an inviolable side constraint to human
action, and individuals may not be used without their
consent for the greater good of others [20].
Some thinkers might argue that actuating the track
switch is not "using" the one to save five since the five

IEEE Technology and Society Magazine

∕

june 2016

http://www.Tech.AD

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - June 2016