The Responsible City: Avoiding Technology’s Undemocratic Social Contracts

Ben Green

1 These days, big tech companies such as Cisco, IBM, and Siemens are extremely eager to
promote their products and services which make a city “smart.” These are, however,
technologies that are likely to bring consequences that are unintended and undesirable.
Therefore, it is important that we ask some questions when developing and adopting these
new technologies. As Langdon Winner (1986) asserts, we need to carefully consider the
social contract implied by building a particular technological system in a particular form. Just
as Plato and Aristotle asked, “What is the best form of political society?”, today we must ask,
Winner argues, “What forms of technology are compatible with the kind of society we want
to build?”. By focusing on the technical and political arrangements of technology, also
known as its architecture, we must ask questions such as “By what means should this
technology achieve its goals?”, “Who should control it?” and “How should we pay for it?”.

2 The answers to these questions can have significant implications. Whether we are aware of it
or not, the technologies we use in smart cities today will play a significant role in defining the
social contract of the next century and shaping the society of the future. Currently, the
architecture of the smart city is fundamentally an undemocratic one: Many technologies
collect unchecked data about individuals and use algorithms that are not transparent. Life
altering decisions can be made relying on these algorithms. Moreover, these technologies
create massive information and power imbalances that favor governments and companies
over those whom they track and analyze. The smart city can be used to increase surveillance,
corporate profits, and social control in secret, which inevitably increases social inequality.

3 The smart city represents the vast expansion of both government and corporate data
collection. Embedding sensors, cameras, software, and an internet connection in everyday
objects from streetlights to trashcans – creating what is known as the “Internet of Things” –
makes it possible to collect remarkably precise data about what is happening in a city. This
data could be used to facilitate beneficial outcomes: reducing traffic, improving
infrastructure, and saving energy. City governments are keen to take advantage of these
new technologies, but they may not always act responsibly to protect fundamental rights.

4 Although well-intentioned, one practice of the New York City government demonstrates how
good intentions can sometimes lead to negative outcomes. In an effort to make the Internet
more accessible to its citizens, the city government and CityBridge formed a public-private
partnership, and LinkNYC kiosks were installed all over the city. Although the program was
initiated with good intentions, there seem to be some undesirable consequences. The kiosks,
equipped with sensors, gather an enormous quantity of data about every device that
connects to the Wi-Fi network: not just its location and operating system but also its MAC
address (a device’s unique identifier that helps it connect to the internet) (CityBridge, 2017).
 Sidewalk Labs, the company that owns and operates the kiosks, claims that this data is
merely “technical information” and not “personally identifiable information” (CityBridge).
This means the data collected does not include information such as names, addresses, or
social security numbers. At first glance, this appears to be a sensible distinction. But just
because data does not contain names, it does not mean it lacks information about people.
Yes, one single data point—a phone’s MAC address at a particular place at a particular time
—is unlikely to reveal someone’s identity or anything sensitive about them. But when
millions of data points are collected, with modern analysis techniques, such data can be used
to track people’s movements and infer intimate details about their lives.

5 Research has demonstrated that it is possible to identify individuals and learn about their
behavior even when data does not include personally identifiable information. When data is
combined with artificial intelligence, it is possible to infer a great deal of personal
information that is not explicitly contained in a dataset. For example, de Montjoye et al.
(2015) analyzed two datasets that contained mobile phone location traces and credit card
transactions of more than one million people – just their anonymous personal IDs, locations,
and times, what is called “merely technical information.” They were able to identify 90
percent of people with just four data points, including information about where they have
been and when they were there. Analyzing detailed information about where you have
been, for instance, machine learning algorithms can even predict whom you know and
where you will go next (Eagle et al., 2009; Backstrom et al., 2010). And this is exactly what
Sidewalk Labs intend to do: obtain more information about people using the data they
collect from the kiosks. The aim is explicitly stated by the company’s founder and CEO, Dan
Doctoroff, who told a public audience in 2016 that the company expects to “make a lot of
money from this” (Pinto, 2016).

6 The collected data in smart cities includes detailed information about the behavior of everyone
within the city, and it will be very easy for municipalities to identify and track individuals,
with worrying implications. Sensors on streetlights and other forms of smart infrastructure,
such as LinkNYC kiosks, can track internet-connected devices in their vicinity, making it
possible to follow people’s movements throughout the city. Similarly, cameras, paired with
software that can identify people or objects, also bring about surveillance threats. In Los
Angeles, for example, automatic license plate readers (ALPRs) record the location of three
million vehicles every week, collecting information that often finds its way into the hands of
U.S. Immigration and Customs Enforcement (ICE) (Henry, 2014; Glaser, 2018). The police-
worn body cameras, which many support as a tool to hold police accountable, also create
the potential for widespread surveillance of all public spaces. Body camera manufacturers
are developing facial recognition software to analyze footage, and only few police
departments in the United States have a policy about body cameras that limit the use of
biometric technologies (“Police Body Worn Cameras,” 2017). Making similar use of
technology, police in Orlando are using Amazon’s facial recognition service to monitor in
real-time everyone who appears in the video feeds from traffic cameras (Kaste, 2018). So the
threat is very much real: the movements of civilians can easily be tracked wherever they go,
participants at protests can easily be identified, and crowds can easily be scanned to follow
individuals who have outstanding warrants (Kofman, 2017).
7 Once these smart city technologies are installed, it is practically impossible to escape being
tracked. Many people point out that there is an option to opt-out: if you don’t like data
about you being collected, don’t use the websites or apps that collect it. Yet, this is not as
simple as it sounds. Since it is almost impossible to communicate, travel, or get hired
without email, search engines, smartphones, and social media, this is an unreasonable
choice for a lot of people. Urban residents face an untenable bind. On the one hand, if they
make the decision to avoid modern technology, they will not be able to follow public
announcements and conversations that occur online. Besides, they will also be losing out on
services that governments distribute by analyzing data (Lerman, 2013). For example, if cities
analyze people’s movements with MAC address sensors to determine where to place bus
stops, they will overlook the needs of those without smartphones and those who turn off
their phones to avoid being tracked. Moreover, in places where cameras are used to identify
people, it is impossible to escape being tracked, even by abandoning one’s personal digital
technology. On the other hand, if they make the decision to use those apps on their
smartphones and other wireless technologies, they must suffer the consequences of being
tracked.

8 Although data collection touches everyone, the most severe impacts of diminishing privacy are
suffered by the urban poor and minorities. Following the LinkNYC case, for example, while
well-off New Yorkers who do not want LinkNYC to track them can forgo free Wi-Fi in favor of
a personal data plan, lower-class residents have no alternative to free Wi-Fi and must accept
being tracked in exchange for internet access. Thus, the inevitable outcome of accepting
pervasive data collection in smart cities is the creation of “a new type of societal system: a
higher class of citizens free from fear of manipulation or control and a lower class of citizens
who must continue to give up their privacy in order to operate within the prevailing
economic system” (Garlick, 2015, para. 10). Thus, smart cities provide welfare offices, police,
employers, data brokers, and others who use data to control the lives of the urban poor with
a new tool for surveillance and exploitation. A single mother could be flagged by an
algorithm and lose welfare benefits because she was identified at a protest by body camera
footage. Also, minorities are prone to suffer the most damaging consequences of being
identified and tracked by the government. Undocumented immigrants may face deportation,
or a black teen could be identified for surveillance by the police because he connects to a
public Wi-Fi beacon that is often used by people with criminal records.

9 The great risks to equity, autonomy, privacy, and social justice introduced by data collection in
smart cities raise new challenges for city governments, and therefore they must act with
great responsibility. Many smart city projects are similar to LinkNYC and involve public–
private partnerships between city governments and private companies. For governments,
working with companies makes it possible to take advantage of private-sector technologies
that would be difficult to develop internally. For technology companies, partnerships with
cities offer a big market for the products and services they sell, and companies need cities
more than cities need companies. Given these dynamics, cities have the opportunity to
assert themselves as market makers, acting both individually and collectively to shape the
direction of smart city technology.

10 City governments must, therefore, thoroughly consider whether the benefits of new services
are worth the price. Should they allow companies to collect untold amounts of data about
the public? If not, they must find ways to obtain the benefits of new technology without
suffering those costs. They must decide what data can be collected and who gets to access it.
Cities are therefore confronted not merely with technical judgments about how to operate
municipal services but with deeply political decisions that will determine the future of urban
life. Will cities increase their control over their inhabitants and provide similar control to
companies without any public dialogue? Or will they ensure that the social contract they
create through technology provides people with a right to the city free from being
monitored and manipulated by corporate and governmental entities? If city governments do
not take action, technology companies may continue to gain opaque and unaccountable
private power over urban life.

Adapted from The Smart Enough City (p. 91), by B. Green, 2019, MIT Press
(https://smartenoughcity.mitpress.mit. edu/pub/yvyv9j2i/release/1). CC BY 4.0
 References

Backstrom, L., Sun, E., & Marlow, C. (2010, April 26-30). Find me if you can: Improving
geographical prediction with social and spatial proximity. Proceedings of the 19th
international conference on world wide web, USA, 61-70.
https://doi.org/10.1145/1772690.1772698

CityBridge. (2017). LinkNYC. https://www.link.nyc/index.html

de Montjoye, Y. A., Radaelli, L., Singh, V. K., & Pentland, A. S. (2015). Unique in the shopping
mall: On the reidentifiability of credit card metadata. Science, 347(6221), 536-539. 10.1126/
science.1256297

Eagle, N., Pentland, A. S., & Lazer, D. (2009). Inferring friendship network structure by using
mobile phone data. Proceedings of the national academy of sciences, 106(36), 15274- 15278.
https://www.pnas.org/content/106/36/15274

Garlick, R. (2015, February 20). Privacy inequality is coming, and it does not look pretty.
Medium. https://medium.com/@rossgarlick/privacy-inequality-is-coming-and-it-does-not-
look-pretty8f8205be55b5

Glaser, A. (2018, March 13). Sanctuary cities are handing ICE a map. Slate. https://slate.
com/technology/2018/03/how-ice-may-be-able-to-access-license-plate-data-from-
sanctuarycities-and-use-it-for-arrests.html

Henry, J. (2014, August 26). Los Angeles police, sheriff’s scan over 3 million license plates a
week. San Gabriel Valley Tribune. https://www.sgvtribune.com/2014/08/26/los-angeles-
policesheriffs-scan-over-3-million-license-plates-a-week/

Kaste, M. (2018). Orlando police testing Amazon’s real-time facial recognition. National
Public Radio. https://www.npr.org/2018/05/22/613115969/orlando-police-testing-amazons-
realtime-facial-recognition

Kofman, A. (2017, March 22). Real-time face recognition threatens to turn cops’ body
cameras into surveillance machines. The Intercept.
https://theintercept.com/2017/03/22/real-timeface-recognition-threatens-to-turn-cops-
body-cameras-into-surveillance-machines/

Lerman, J. (2013, September). Big data and its exclusions. Stanford Law Review Online.
https://www.stanfordlawreview.org/online/privacy-and-big-data-big-data-and-its-
exclusions/
Pinto, N. (2016, July 6). Google is transforming NYC’s payphones into a ‘personalized
propaganda engine’. Village Voice. https://www.villagevoice.com/2016/07/06/google-is-
transformingnycs-payphones-into-a-personalized-propaganda-engine/.

Police Body Worn Cameras. (2017, November). The Leadership Conference on Civil and
Human Rights and Upturn. https://www.bwcscorecard.org/

Winner, L. (1986). The whale and the reactor: A search for limits in an age of high
technology. University of Chicago Press.