The Challenge of Machines in the 21st Century
Information technology and the internet are changing the way democracy works.
Recent revelations of the use of personal data to manipulate elections tell us that we live in a very different place we thought we did just weeks ago. Marketing companies, like the now infamous Cambridge Analytica, may deploy data profiling to influence human targets on social media. This involves the enveloping of the subjects within an artificial world; Christopher Wylie, the whistleblower at the center of this scandal, referred to these worlds as “cultures.” In each of these artificial cultures, political candidates would appear to each target from a different aspect, but always as a perfect candidate tailored to the psychographic profile of that particular voter. This approach, Cambridge Analytica claims, would increment the candidate’s electoral margins.
There is currently no information if the use of personal data had a deciding effect on the US presidential elections. However, the process is revealing of the power online companies hold today to, in principle, manipulate its customers.
Technology envelops the modern citizen, as the online space invades everyday living and most of one’s material existence. Even the very notion of self is drastically changing. Being cognitively stimulating, this virtual reality absorbs time, energy and capital. Mankind transformed the internet into a concrete repository for what the Austrian philosopher Karl Popper called “world 3” patrimony — the products of the human mind, such as stories, theories, songs and languages.
The internet now stores a huge chunk of human knowledge and this repository is getting bigger everyday. Some even claim that humans, in a few years, will quickly build a new form of civilization, in which Artificial Intelligence (AI), supported by the all the knowledge available online, will quickly surpass individual human intelligence. The change has been so fast and furious that very few would dare to predict its consequences and implications, even in the short term.
The Global War for Attention
One of the main characteristics of this digital universe is its ability to collect phenomenal amounts of data. The ethical consequences of this process are still significantly unknown. Collecting personal data in order to build artificial worlds around political candidates may be extremely dangerous to democracy. The informed democratic decision-making process implies that the citizen should have access to the best information. Unequal knowledge, as has been shown, often leads to bad decision-making. In the case of markets, it drives away quality. In politics, when someone votes based on fake information, he or she is making the wrong choice. And good candidates tend to avoid vitiated games. When a company decides to transform the rules of political propaganda, things can take unexpected and unwelcome turns.
The current abuse of digital surveillance and the collection of online usage information by big companies may effectively violate consumer rights. One good example is the pricing manipulation based on browsing history or location data: On some websites you will get a different price if you visit the webpage for the same product twice. We need to change the conditions in which technology now operates. At this point, we won’t be able to stop technological evolution, but its influence on our daily lives should be carefully examined, supervised and regulated, in accordance with sound ethical principles. Many companies are already aware of these emerging issues.
Online companies follow cold profit-seeking logic of attention capture. The non-excludable nature of information — the property of hardly being exclusive and eventually being copied at zero cost — has lead to a global war for user attention. The digital economy is now, most of all, an economy of attention, which is now a scarce resource. There is no reason to believe that products created to capture attention, like big box-office hits and popular TV-shows that generate massive profits, are the same that consumers would wish to see produced. When it comes to overall quality of the mass media production, we are living in the age of the crowd. If media producers nurture cultural diversity, it is because it drives innovation. As a disproportionate share of big media events makes obvious, cultural mainstream products lack the quality and the critical nuance of minority products.
Value of Information
It is impossible to determine the value of information products before consumption, so consumers need assistance in the buying process. This assistance has been leveraged by the power of recommendation systems that bring voting processes — liking what had already been consumed, correlation patterns between similar products, trend analysis and intense usage surveillance into play. In this asymmetric information scenario — in which platforms have the knowledge of many different user preferences and the preferences of each individual user — companies have the ability to know much more about the consumer, in the particular context of buying and bargaining between competition, than the consumer him or herself. It is not a fair game.
We may separate the data that is processed online into two distinct categories. There is personal data that uniquely identifies the physical person, such as movies, photos and verified user information. This type of data is mostly present on social sites. Then there is usage data resulting from the human computer interaction: click records, browsing history, computer cookies and other device specific signals like GPS data. The purpose of this data is mostly to enhance the usability experience, but it can also be used for purposes that are not immediately evident, such as identifying a specific user by his or her style of typing.
The European Union’s General Data Protection Regulation that comes into force on May 25 will surely bring new limits to the abuse of personal data. The owner of the information will have more responsibilities, and hopefully the so called “data subject” will have his rights more broadly protected. The new regulation extents its territorial scope, penalties and conditions, meaning EU citizens will be protected regardless whether the processing of the data takes place in the Europe or beyond. The penalties will be harsher, and privacy protection is becoming an integral part of system design. The effect on big social platforms like Facebook could be profound. In case of abuse, Facebook may be forced to cooperate much more extensively that it does now. However, the second type of data is still unregulated, and will probably remain so for the time being.
Although personal data may have a particular commercial interest, the data generated by user interaction can have a personal impact because it can reveal much more about the user than he or she suspects. Location data, browsing history and click patterns may be correlated and compared with already compiled aggregated user knowledge in order to build user profiles. Automatic editing of user timelines, like Facebook and Tweeter newsfeeds, can change the way each user perceives his corner of his favorite social network and the virtual part of his or her world. The Google Gold Mine is supported on clever automatic analysis of usage data. When aggregated, user data can become even more valuable. Large scale aggregated social data can become a precious tool for future public policy design. However, this particular purpose will determine the good this may bring.
Voice of the Crowd
Data constitutes a fundamental pillar of modern science and of progress. Without data, scientist could not test their hypothesis against the crude reality and then obtain replicable truths. If this is important for the natural sciences, it is much more relevant for the social sciences. Until very recently, social scientists could only rely on small snapshots of social phenomena, like personal interviews and focal groups. With the advent of the internet, the possibility of gathering vast amounts of perfectly structured social data extended the horizons of social sciences enormously. Now the social scientist can hear the voice of the whole crowd.
However, the gigantic cauldron of generated online information is still unregulated, so use of data should become subject to serious regulation. First of all, we need to decide if any sort of usage data should be stored. Then there should be limits for its application in the configuration of device interfaces. What is the benefit of just seeing the content social sites automatically choose according to online patterns of use? Should we expect to see automatic personalized real-time news selection in the future? If yes, then we should prepare ourselves for the next big technological leap. How much should we allow AI devices to know about ourselves?
Owning information given by users and consumers in this economy of attention constitutes an ethical responsibility. Most online platforms today become indistinguishable in many aspects from our physical reality, they are user-friendly and cognitively savvy. Society must devise ethics rules for machines, so that the fast transition from the physical world to the immaterial world doesn’t destroy the most dearly-held human values.
The views expressed in this article are the author’s own and do not necessarily reflect Fair Observer’s editorial policy.