IEEE Technology and Society Magazine - Spring 2013 - 61

eacting to the turbulent events of the
last decade or so,
the U.S. government
has focused much
attention on developing strategies
to curtail the misuse of technology, especially biological agents.
In principle, any technology can be
exploited and used to achieve malicious ends. This is an especially
poignant issue now that we have
reached the digital age and that
an individual person's actions can
have an impact that spans across
the globe. One way in which a
malicious actor could cause serious harm is by manipulating how a
robot functions.
robotic technology has been used
for quite some time by the military
and in industrial settings, and robots
are making their way into our homes
as well [1], [2]. Considering the magnitude of harm that could result if
a robot is misused, we contend that
networked robots should be a particular source of concern. There are
many types of networked robots that
are already in existence and even
more that are on the horizon, including robotic swarms and aerial drones
that can communicate with one
another. However, our primary focus
here will be on a subset of networked
robots, those that are designed to be
connected to the Internet. Our intent
is to emphasize the serious ethical
responsibilities that professionals
have while they are engaged in the
design, development, and deployment of robots. These responsibilities are already important, but they
will become increasingly so as
robots become more complex and as
they become more highly integrated
into our lives.

Vulnerability in
the Digital Age
For many of us, day-to-day tasks
would be nearly impossible to
accomplish without a computer or
other electronic devices. As Henry
petroski states, "For all their idiosyncrasies and faults-and features

known and unknown-we trust
computers, as we do our confidants,
with our most private thoughts and
count on them for our most critical
financial calculations" [3]. Yet this
reliance can make us open to a variety of different harms, including
accidents, increasingly risky behaviors, and intentional malicious acts.
The ways in which computer vulnerabilities can be exploited range
widely from the fairly innocuous
to the extremely serious. For example, the "peace virus" that infected
Microsoft Excel spreadsheets years
ago may have been a nuisance, but
it did not seem to cause lasting
harm [4]. An example of a more
serious event is when hackers were
able to access systems used by the
U.S. military [5].
Even after an electronic security
issue is revealed publicly, it can be
difficult to eradicate. The group
"Gotsec Security"1 has gained
notoriety (and attention from
international law enforcement) for
discovering and exploiting security vulnerabilities. Further, it is
difficult to anticipate the myriad
vulnerabilities that a particular
technology may have. Hackers
have been able to redirect computer
users to the wrong Internet servers
in several different ways [6]. In
2011, the U.S. FBI identified, and
Estonian police arrested, members of a group who had infected
thousands of computers worldwide
[7]. When an infected computer
sought an Internet UrL, the infection directed the request to a compromised DNS server. Even after
the scheme was identified and the
perpetrators arrested, the infection
continues to be a problem. In unraveling the damage, the FBI took
down the compromised DNS servers, replacing them with benign,
FBI-controlled servers until the
infected machines could be purged
of the infection. Although the measure has been discontinued, the
FBI had been running these servers
1

http://security.goatse.fr/, accessed Oct. 12, 2012.

IEEE TECHNOLOGY AND SOCIETY MAGAZINE

SprING 2013

for months; the FBI delayed taking them down because not all the
infected machines have been fixed,
and the "temporary" servers were
still being used [8].
Electronic devices have numerous weaknesses that can be
exploited, and some electronic
attacks are rather subtle. For
example, information typed into a
computer keyboard can allegedly
be gleaned from the keyboard's
cables [9]. reports suggest that
hackers have the capability to tamper with mobile phone networks
[10] and with the Sony playstation [11]. A case in Austin, TX,
made the public aware that hackers
could reprogram street signs [12].
Although the action did not appear
to be maliciously motivated, it does
illustrate the point that common
electronic devices and systems are
susceptible to attack. Transmitters can change the light color on
traffic signals, which opens the
door to the potential for abuse [13].
Further, researchers suggest that
"GpS spoofing" could be used to
compromise numerous kinds of
electronic technologies, including
aerial drones [14], [15].
Robots and Vulnerability
Now that robots are becoming
increasingly common, it is time
to more seriously consider possibilities for harm due to the technology. The first reported case of
a robot taking the life of a human
is several decades old now [16]. It
is important to note that the accident occurred in an industrial setting because the behavior of this
kind of robot is probably much
more predictable than the behavior of the numerous categories of
robots that will enter our lives in
the near future. An industrial robot
often operates in a fixed location
and performs the same, repetitive
act. Forthcoming generations of
robots, including socially assistive ones [17], will be integrated
into a large variety of, and often
dynamic, contexts where it will be
|

Table of Contents for the Digital Edition of IEEE Technology and Society Magazine - Spring 2013