Literature review diagram for progress of artificial intelligence and review of couple of articles

FEATURED FORUM: WELCOME TO THE DIGITAL ERA: THE IMPACT OF AI ON BUSINESS AND

SOCIETY

Ethical Aspects of the Impact of AI: the Status of Humans in the Era of Artificial Intelligence

Roman Rakowski1 & Petr Polak2 & Petra Kowalikova1

Accepted: 6 May 2021 # Springer Science+Business Media, LLC, part of Springer Nature 2021

Abstract On the one hand, AI is a functional tool for emancipating people from routine work tasks, thus expanding the possibilities of their self-realization and the utilization of individual interests and aspirations through more meaningful spending of time. On the other hand, there are undisputable risks associated with excessive machine autonomy and limited human control, based on the insufficient ability to monitor the performance of these systems and to prevent errors or damage (Floridi et al. Minds & Machines 28, 689–707, 2018). In connection with the use of ethical principles in the research and development of artificial intelligence, the question of the social control of science and technology opens out into an analysis of the opportunities and risks that technological progress can mean for security, democracy, environmental sustainability, social ties and community life, value systems, etc. For this reason, it is necessary to identify and analyse the aspects of artificial intelligence that could have the most significant impact on society. The present text is focused on the application of artificial intelligence in the context of the market and service sector, and the related process of exclusion of people from the development, production and distribution of goods and services. Should the application of artificial intelligence be subject to value frameworks, or can the application of AI be sufficiently regulated by the market on its own?

Keywords AI . Big data . Datafication . Commodification of data . Digital ideology . Ethical aspects

Introduction

We live in a period of digital turn, which is often referred to by media, theorists and experts as the fourth industrial revolution or Industry 4.0. The 4.0 concept was originally intended in relation to the field of industry and production, in which there will be such great changes that the whole social sphere will subsequently change – as was the case during previous tech- nological revolutions. The opposite is true; it is necessary to

talk more about the inconspicuous technological evolution that is taking place at all levels of society, not just at the level of the industry. The reach of modern technology has long gone beyond research, development and manufacturing and has completely dominated public and private life to the point that 4.0 seems to be a society based on the interconnection of technology, people and data (Big Data). However, this means that new ethical and political challenges lie in the implemen- tation of new technologies. On the one hand, technologies are radically changing the environment in which we live, and on the other hand, without us realizing it, they are also changing ourselves. In the context of the “digital turn”, a transforma- tion is currently affecting established modern oppositions such as subject/object, public/private, consumption/produc- tion, mind/body, work/leisure, culture/nature and so on (Chandler & Fuchs, 2019, p. 2).

The initial enthusiasm for scientific discoveries and inno- vations is seldom marked by fears of the unintended conse- quences of their practical application. The obstacles consid- ered include the restriction of the field of application by leg- islative standards or the economic aspects of the transposition

* Petr Polak [email protected]

Roman Rakowski [email protected]

Petra Kowalikova [email protected]

1 VSB - Technical University of Ostrava, Ostrava, Czech Republic 2 Faculty of Business and Economics, Mendel University in Brno,

Brno, Czech Republic

https://doi.org/10.1007/s12115-021-00586-8

/ Published online: 26 May 2021

Society (2021) 58:196–203

of new technology from laboratory conditions into production practice. The existence of some difference between technolog- ical possibilities and their implementation in an environment limited by economic, legal, and organizational factors is wide- ly accepted. However, a silent precondition for the introduc- tion of technological innovations is their presumed benefit for individuals, social groups, or society as a whole. Possible negative consequences remain below the threshold of discrim- ination, provided that they do not directly conflict with bind- ing legislative or social standards in general and can be offset by positive effects in the relevant area. However, the more rapid the technological development and the more important the social role that new technologies play, the more carefully their impacts on an individual’s life and the functioning of individual social subsystems should be considered (Matochova et al., 2019, p. 229; Kowalikova et al., 2020, pp. 631–636).

Responses to the dynamics of current change range from attempts to stabilize the environment by introducing new con- trol mechanisms and increasing the frequency of controls, to the adoption of change and restructuring of hitherto known interpretation schemes, to feelings of helplessness and alien- ation (Veitas & Weinbaum, 2017, pp. 1–2).

The constant presentation of risks in public space and the constant effort to reduce them significantly contribute to the disruption of the feeling of ontological security. Compared to previous stages of social development, advanced societies are now more likely to die from overeating rather than famine, suicide rather than an attack by soldiers, terrorists or criminals, and of old age rather than an infectious dis- ease (Harari, 2018, p. 397).

The American theorist and philosopher Fredric Jameson, in his famous book Postmodernism or, The Cultural Logic of Late Capitalism (Jameson, 1992), argues that new technolo- gies help shape the subject itself under the weight of late capitalism (which is denoted by the term postmodernism). Literally Jameson, in line with the Kantian interpretation of aesthetics, speaks of the technological sublime as something we are not able to reflect on from our position and understand at all (cognitive mapping). Although this thesis is particularly concerned with the periodization of postmodernism, there is another assumption in Jameson’s theory that is important for understanding people in the world of new technologies (espe- cially algorithms, AI, big data). This is a certain transforma- tion of a social subject that adapts quickly to new “postmod- ern” trends (change in the dynamics of the relationship be- tween culture and economy, the emergence of new services, the transition to digital capitalism). If we take this analogy out of the context of the 90s and insert it into the present – the time that shows signs of a technological turnaround – it can be assumed that the syntax of the times is an algorithm applied to big data, which is mediated by new technologies (which in turn are the medium of new services and business models).

These algorithms then “help” us in orienting ourselves with the inexhaustible amount of data that new technologies (in- cluding information and communication technologies) pro- duce ad infinitum (Ross, 2017). However, the design of these algorithms and of artificial intelligence is not neutral and hides certain pitfalls in the form of ideologies or biases that are not easy to decode (Bowles, 2018).

Big Data is an integral backdrop of our lives. However, it is useless to us if we cannot employ it in real time in the form of personalization of various services. It decides what movies we will watch, what music to listen to, where we go on a trip, where we stay or whom we meet and whether we get a mort- gage, whether a package from Amazon will arrive at our ad- dress or whether our device’s camera gives us access to our notebook based on our race (Bridle, pp. 142–143). Selected camps of theorists in such cases do not waste a moment to use the term technological determinism, which points to the au- tonomy of technology. However, we will try to go beyond this pessimistic approach in this study.

Adam Greenfield’s book Radical Technologies: The Design of Everyday Life (Greenfield, 2017) offers an interest- ing depiction in this pessimistic context. Let us imagine that we are sitting in a café recommended by an algorithm; we pay for the coffee in cryptocurrency via a smartphone, while chil- dren across the street play AR games on smart devices. This would not have been possible at all a few years ago, but today it is understood as a common routine. The whole situation is drawn up by technologies, however, not with one technology but rather with a set of individual technologies and services. At first glance, it may seem to us that these technologies are too separate to be functional and create this situation. However, their advantage is that they can be connected by an “interface” of ones and zeros. This also multiplies the effi- ciency of individual technologies. (Greenfield, 2017, pp. 498– 500). However, it is clear that this mediation between different technologies – as we will show below – needs a clearer inter- pretive framework.

If we take this technological allegory to the extreme, we can say that technologies can to some extent constitute social reality (remember the social bubbles on social networks, the degradation of public space – new agoras). In updating Jameson’s theory and with an inclination to the (problematic) technological determinism, we could say that the subject in the technological turn adapts to the syntax of algorithms, artificial intelligence and Big Data. Social reality can then be deprived of chance and subtlety to the extent that it makes it seem that it can be transformed into the formal lan- guage of ones and zeros. The aforementioned Bridle also looks at this issue highly pessimistically: “In this way, com- putation does not merely govern our actions in the present, but constructs a future that best fits its parameters. That which is possible becomes that which is computable. That which is hard to quantify and difficult to model, that which has not

197Soc (2021) 58:196–203

been seen before or which does not map onto established patterns, that which is uncertain or ambiguous, is excluded from the field of possible futures” (Bridle, 2019, pp. 44).

The fact that adds to this pessimistic view of mankind is that we ourselves have ceased to perceive algorithms and new technologies as constructs of our everyday reality. Bridle thus points to a problem that can be illustrated in the philosophical direction of functionalism. Every day we use the outputs of new technologies, but we have no idea how they work and what algorithms are hidden in programs, services and advertisements. Without understanding the con- sequences, we use these technologies as black boxes, as functions in which we enter and receive data (Bridle, 2019). This problem is illustrated by simply scrolling on social networks: the posts we see are already preselected to get our attention. If we had to see all the posts of all our friends, for example, on Facebook, we could roll for hours before seeing something that really interests us. We could thus claim that the emancipation program of the Enlightenment is unfinished in this case, because in the technological turn, we unknowingly leave most of the deci- sions to the algorithms and artificial intelligence. Algorithms do not even have to work too much; social reality is simpli- fied to the level of formal language. And that is the reason why algorithms can have such an effect. It is therefore better to not look for complexities in algorithms but rather at the simplicity of social reality. Our social reality is complex and diverse, but due to algorithms, it is no longer random. This is the world of computational hegemony. However, the question remains how to prevent this: responsibility and rules (ethics) or awareness and education (breaking ideology)?

In connection with the possibilities of using AI, Makridakis (2017, pp. 8–11) presents four ways of interpreting the im- pacts of this technology on the functioning of society. Optimists predict the utilization of the speed and memory capacity of computers and the ability to share their knowledge with the human brain. Technological innovations will allow genetics to intervene in the genetic code to prevent disease, ageing or even death. Nanotechnology will make it possible to create virtually any product at low cost, and robots that will take over all human work will allow people to choose their way of spending their free time and to choose work activities according to their interests. Pragmatists rely on the ability to control AI through effective regulation. Rather than on AI which seeks to mimic human intelligence, they focus on AI’s ability to expand human capabilities to increase room for human decision-making and control. Doubters deny dys- topian scenarios based on the threat of AI, pointing out that human intelligence cannot be replicated and captured in a set of formal rules. And if so, even then it will not be possible to machine-replace human creativity, which is based on overstepping rules – on antialgorithmic behaviour. It is

creativity, which is based on the violation of established norms and ways of thinking, that also other authors (e.g. Jankel, 2015) consider to be an ability non-replicable by computers.

Harari (2018) warns against the division of intelligence and consciousness and draws attention to the potential danger of using unconscious but highly intelligent algorithms. If we accept the assumption that organisms are algorithms and life is data processing, then humans cannot compete with a ma- chine that is able to make decisions based on the evaluation of all available information and process a problem situation with- out consciousness – or precisely because of the lack of it – with a better result. A simple example is the comparison of accidents between autonomous vehicles and people-driven vehicles. The mass expansion of these types of vehicles would result in a significant increase in unemployment among pro- fessional drivers. Which would, in a sense, confirm the supe- riority of the machine over man (Makridakis, 2017, p. 10). After all, dystopian visions assume that sooner or later, orig- inally human-made decisions will be dominated by the ever more perfect machines, with better results than people prone to errors would be able to achieve. This would necessarily change the whole system of social stratification. Exclusion or reduction of the role of a person in key decision-making processes connected with the functioning of society would then necessarily lead to their inferior social status.

Impacts of AI Use on Business and the Labour Market

The determining influence of digitalization and automation on the functioning of society at the level of all social subsystems is indisputable. For the destructive and creative impacts of digitalization and informatization on the labour market (crea- tion, extinction and transformation of professions or jobs), there are also proposals for political, economic and social measures (increase of the minimum wage, introduction of un- conditional basic income, support for the elderly and the low- skilled, etc.). Changes in the structure of the labour market must be accompanied by radical structural changes in society and in the way people think about work.

What has become the subject of analyses are the social consequences of technological innovations, issues of social control of science and technology with a special focus on opportunities and risks that technological progress can mean for social ties, political life, value systems, etc.

The positive social potential of artificial intelligence can manifest itself at the level of supporting and securing the func- tioning of key subsystems of society without unspoken ste- reotypes, prejudices and hidden discriminatory behaviour. Furthermore, it may be reflected in changes in the structure of work-related and non-work time due to the reduction of

198 Soc (2021) 58:196–203

activities performed by people and thus in the expansion of space for social activities (development of social relations, community life, volunteering, etc.). At the same time, possible negative impacts of the development, implementation and ex- pansion of AI at the economic, political and social levels are considered (insufficient sociopolitical reflection on changes in the structure of the labour market, misuse of AI by nondem- ocratic regimes, limited possibilities of AI control, etc.). In this context, questions arise as to who and how should be involved in the decision-making in the development and implementa- tion of innovation; on the basis of which criteria states should set priorities for R&D funding; how companies should mea- sure risks and set safety standards; whether and how experts are obliged to communicate to the public their decisions and their reasoning; etc. (Matochova et al., 2019, pp. 230–231).

Floridi et al. (2018, p. 690–694) emphasize the possible use of artificial intelligence technology to support human nature and its possibilities. On the one hand, he considers AI to be a tool for expanding the possibilities of individual self- realization and the utilization of interests, abilities, skills and aspirations. Mastering routine tasks through AI opens space for more meaningful ways of spending time. On the one hand, Floridi points out the positive use of advanced intelligence in human decision-making and action. On the other hand, he draws attention to the necessary responsibility in the develop- ment and distribution of state-of-the-art technologies, which should remain under human control and benefit all members of society as fairly as possible. AI technology enables more efficient functioning of society and social systems, from the prevention and treatment of diseases to the optimization of transport and logistics to a more efficient redistribution of resources or a more sustainable approach to consumption. However, the power of technology also brings the risks of its use. According to Floridi, these are mainly associated with excessive autonomy of machines and limited human control, based on insufficient ability to monitor the performance of these systems and prevent errors or damage. A balance needs to be struck between the ambitious projects and opportunities that AI offers to improve human life and the strength of the control mechanisms that people and societies set up.

Hawksworth et al. (2018, pp. 1–17) in their report identify three phases of AI involvement in the functioning of various areas of society but especially with regard to the shape of the labour market. Until the early 2020s, they expect an algorith- mic wave that is reflected in the automation of simple com- putational tasks and the analysis of structured data. For this reason, they consider the sectors based on routine data pro- cessing, i.e. finances and insurance, but also the area of infor- mation processing and communication, as the most accessible to automation – and most risky in terms of maintaining the number of jobs. The second half of the 2020s will be hit by a wave of augmentation based on dynamic interaction with technology in administrative support and decision-making

and on the automation of repetitive tasks, including the anal- ysis of unstructured data in partially controlled environments. The sectors concerned will be public administration and self- government, production, warehousing and transportation. In the 2030s, the autonomous wave should reach its peak, which presupposes full automation of physical labour, machines with manual dexterity and problem-solving skills in dynamic situations and in the real-world environment, where an imme- diate response is required. This phase of the use of state-of- the-art technology will affect the construction sector, water management, wastewater treatment and waste management, etc.

In their analysis, Hawksworth et al. (2018, p. 2) assume that in the short term, the most vulnerable jobs will be in the financial sector and insurance and jobs held more frequently by women. From a long-term perspective, the vulnerable group is represented by employees in the transportation sector, rather than men and people with lower qualifications (which confirms the importance of investing in lifelong learning or retraining). The same authors identify risk areas in terms of the negative impact of the automation process by country, indus- try and type of worker. The share of jobs at risk reflects the country’s average level of education. It thus ranges from 20 to 25% of positions in some East Asian and Nordic economies with a high level of education of the population to 40% of positions in Eastern European economies, based mainly on industrial production. Among these extremes are economies dependent primarily on services, but with a significant propor- tion of low-skilled workers (UK, USA). Within 10 years, after the widespread use of autonomous vehicles, transportation will be one of the most vulnerable sectors in terms of main- taining the structure of jobs. Currently, the riskiest sectors are those dependent on routine processing of structured data such as the financial sector and insurance. At the same time, the least vulnerable groups include workers with a university de- gree who, in addition to their expertise, also show a higher degree of adaptability to technological change. Such qualified employees are also more likely to hold higher management positions, where a lower level of susceptibility to automation is expected. Like actors with lower education, older workers may have a lower degree of adaptability. In the case of manual work and positions in the transportation sector, where men are more frequently represented, a higher degree of threat to the stability of positions can be assumed again through the pro- cess of automation. However, the same is true for women in administrative positions.

Ethics of AI vs. Ideology of AI

The book Future Ethics, together with the theory of Andrew Feinberg, points out that technologies are not inherently neu- tral (Bowles, 2018, pp. 2), and an ideology is encoded in their

199Soc (2021) 58:196–203

very design to distort our usage: if technology forces us to pay attention to it, that was the intention.

Design is applied ethics. Sometimes this connection is ob- vious: if you design razor wire, you are saying that anyone who tries to contravene someone else’s right to private property should be injured. But whatever the medium or material, every act of design is a statement about the future. Design changes how we see the world and how we can act within it; design turns beliefs about how we should live into objects and envi- ronments people will use and inhabit. (Bowles, 2018, pp. 4).

However, the problem that Bowles outlines here can be included in the normative level, where he works with three levels of ethics (deontological ethics, ethics of virtue, utilitar- ianism) and leaves it to the designer (of AI and algorithms) to decide ethically – whether the consumer succumbs to the ide- ology of design is purely up to them, the problem of ideology is transferred to become their responsibility. In essence, this is a naive normative guide for the digital capitalism industry. The problem, however, is that Bowles excludes those that are most affected – the technology company/technology users and data producers – from decision-making. In this context, it is clear that the political theory of technology needs to be thought about rather than ethics. The political theory of tech- nology offers an opportunity to change it – the democratiza- tion of technology, that is, how to intervene in its design – retroactively through society. We can see this possibility on two levels: (1) deideologization of technology – one must realize that we can actually influence technology by our deci- sions (Allmer, 2017); (2) democratization of technology – through a clear disagreement or detournement of technology, we can achieve a change in the goal of technology (Feendber, 2009).

On one side, there is a responsible designer, on the other a conscious society. Should the designer succumb to the values and ethics of the company, there is a society that is being used by the company. It is therefore clear that the requirement of ethics alone is inefficient; the competitive environment itself would have to change.

Ethics and Political Philosophy of AI

If we want to talk about political philosophy and the ethics of artificial intelligence, we should distinguish between political philosophy and ethics – albeit inextricably linked – in relation to new technologies. If we look at the ethics of AI, the most common approaches that appear in the context of the algo- rithm are deontological ethics, ethics of virtue, and utilitarian- ism. Rather, we are talking about the individual level, where the design itself is produced, which is supposed to have a certain impact on the individual and society. However, if we look at the political philosophy of new technologies (algo- rithms and artificial intelligence), we should inquire more

broadly into whether the new technologies concern society as a whole. Here we then should distinguish between the (A) critical and the (B) liberal branch of the political philosophy of technology. (A) The critical theory of technology looks at the power and ideological relations of technology – as in material capitalism, new technologies are considered only as means of production. Here, the specific contradictions that lead to the non-transparent design of AI (how one is deliberately manip- ulated by data in favour of digital capitalism) should be theo- rized. (B) We should ask how to set the rules, on the one hand, so that technologies are not too limited/regulated (the issue of freedom) and, on the other hand, so that these technologies are created for the benefit of society and the problems of the current environmental crisis. These are purely political issues of technology.

Critical Theory of Big Data

In his theory, Allmer (2017) works with tools that allow shared data to be critically examined from the perspective of economic-power relations. Although data seems to be handled by users, the data is actually owned by large companies, which ultimately decide how to handle it. However, such a fact is worrying, and it is necessary to examine the extent to which it affects the user (i.e. the social entity).

The main premise is that capital is accumulated through user data, making this digital environment (such as social media) an arena of struggle in which (as in any production mode) class and social contradictions arise (Allmer, 2017, pp. 5). The fact that this principle of capital accumulation has been transferred from the material environment of commodities to the digital world is part of the evolution of commodification. Commodifying public goods (such as data) has a number of complications: digital reproduction emphasizes the privatiza- tion of data. For this reason, it is necessary to create new forms of capital, and it is best to involve the very user, who is con- stantly producing data, in this digital production. If we stick to the vocabulary of critical theory, this phenomenon can be labelled by the terms of digital alienation and digital exploitation.

For this analysis, Allmer uses Marx’s reasoning, which he places in the current context. According to such an interpretive framework, in a capitalist society, the asymmetry of the rela- tionship of power is embodied in the very design of technol- ogy (Allmer, 2017, pp. 16–26). Technology is understood as a reflection of social relations, and for this reason, as we have seen above, it cannot be understood as neutral. The goals of technology thus correspond to the goals of capital itself (Ibid.). Thus, the technology cannot be designed outside of a social context. The illustration of these theses can be depicted as the birth of a new rationality, which comes with technology at the time of industrialisation and is the essence of mass production

200 Soc (2021) 58:196–203

and the transformation of the whole so-called base (Horkheimer & Adorno, 2007).

The problem with rationalization is that if technology can be taken out of context (e.g., historical expropriation), the essence of rationalization will still remain in it – for example, the question of automation does not lead to human emancipa- tion (as was originally the idea of Herbert Marcuse). The problem is how to work with the potential of new technolo- gies. How do we even discover the emancipation potential of new technologies? Are technological or political changes needed for this emancipation? In this case, we will be helped by the critical theory of technology (dialectics of technology and society), which points to the socially conditioned con- struction of technology and the impact of technology on soci- ety (Allmer, 2017, pp. 42).

Democratizacion of New Technologies and Big Data

Following the example of Feendber, 2009), one can distinguish two main currents in the theory of technology. The first is the so-called instrumental, which speaks of technology as the inter- connection of technology with the value context of society (culture or politics). Technical tools are understood as neutral means serving only social goals. Technology is just a tool to achieve efficiency. Such an approach is purely functional. Technology is designed outside of political ideology. The sec- ond stream, the substantive one, attributes autonomous force to technology that prevails over traditional and competitive values. It, therefore, denies the neutrality of technology and emphasizes the negative consequences of technology for hu- manity and nature. Technology has become part of the lifestyle and everyday life. Technology has dominance over us and there is no escape from it. The opposite is a return to the traditional values of romantic simplicity (a certain apocalyptic vision).

If we think about the ethics of AI – that is, that the designer can consciously modify the ideology of an algorithm and AI – we should also think about the defence options of society, which will be affected by AI. As we saw earlier, Feenberg’s theory of democratization of technology could help us with this.

Feenberg’s theory represents a non-deterministic approach to technology. Technology must not be considered as a set of devices or the sum of rational goals; that would be too func- tionalist. In relation to society, technology must be interpreted like any other artefact. If we overlook the connection between society and technology, then we will perceive it as self-produc- tive. However, technology is political; it is not born in a vacu- um, outside of political ideology, but always in a specific social discourse. For this reason, Feenberg also continues the tradition of critical theory and assumes that public opinion will interfere

with the nature of technology, i.e. the normative requirement of democratic instrumentalization (Feendberg, 2009, pp. 146). Technology is behind political and economic power, and for this reason, it should be part of the public space debate: we cannot be civically autonomous if we do not have the opportu- nity to decide on this industrial process. However, Feenberg realizes that communism has failed to meet this demand (Feenberg, 2014, p. 708). And this is exactly the reason to study technology at all. It seems the self-propagation of innovation prevails over the rational use of technology. Should we, as citizens with rights, not be able to decide what technologies will or will not be implemented in society? This also applies to the current technological turn and Big Data.

Specific values are regularly embedded in technologies – albeit unknowingly – and it is the hermeneutics of technolo- gies that we should use to interpret them.

Technology then shapes the principles we live by. Technologies can to some extent represent our interests (if I own a car, it represents my mobility; if I prefer cycling to work instead of a car, it represents a value). Then we may ask, do these interests define society? However, there is a tendency to look at technology and politics separately. Technological design should be value-oriented rather than market-based instrumentalization.

Technology is thus not a neutral tool: it has its own value, but at the same time society can determine the direction of its development. Technology is the result of many factors: the meaning of technology is defined only by its use in the context of society. Progress alone cannot assign purpose to technolo- gy. This is where Allmer’s inspiration comes from: on the one hand, there is a need to analyse the ideology of how people have a false knowledge that they can influence technology and at the same time the need to democratize technology for eman- cipation. We should ask how theoretical normativity can be implemented in the political functioning of technology.

According to Feenberg, the demands of democratization can come into practice through two mechanisms: the techno- logical code or the democratization of technologies through initiatives that can gradually change the legal framework. We should then recode it to the question of whether technology can help us, for example, in an ecological crisis (updating current positions on technology: on the one hand, socialist technology helps to emancipate people, but on the other hand it is not ecological). However, the question remains, how to involve the widest possible public in technology decisions when technology development is distributed completely inde- pendently of society through autonomous agents?

The democratization of technology states that technology will be an aspect of public life (we are already familiar with civil juries, discussions, protests, boycotts, etc.). This is im- portant at the moment when AI is entering all areas of indus- try, healthcare, services or social relations. In these aspects, however, it is difficult to talk about any regulation of

201Soc (2021) 58:196–203

technology at the expense of freedom. Neither the scientific nor the private sector can be planned or interfered with. For this reason, we are talking about AI ethics and the political philosophy of new technologies. In this context, many con- temporary experts talk about rehabilitating the concept of the social contract in the context of the digital turn: the question of how to democratically shift technological design to social, not instrumental values, without restricting the market freedom of individual actors.

Conclusion: The Social Contract, AI and Digital Turn

The concept of the social contract speaks of an imaginary contract that people have concluded between themselves to ensure a certain area of freedom and social security (the op- posite is the chaotic so-called natural state or homo homini lupus). This concept is very useful for politics and ethics. The rehabilitation of this concept in Rawls’s thought experi- ment called the veil of ignorance (Rawls, 1999) can be applied to the digital turn. People agree on rules that are essentially egalitarian. However, this is not observed in the market envi- ronment as to not lose profit. People have no idea that new technologies deliberately avoid the social contract. The social contract only works if all participants agree to its terms. Bowles believes that this concept can be applied to a technol- ogy society in which this fair system could be applied. “Beneath a veil of ignorance, we wouldn’t know our social status, our intelligence, or even our interests; but if the system is fair we should be satisfied wherever we ended up” (Bowles, 2018, pp. 56). The veil of ignorance is related to deontological ethics – no one would want to take down the veil and find out that they are just a means to an economic goal. The veil of ignorance thus forces us to take into account all the roles that appear in the system – if we create a system of persuasion (e.g. algorithms and AI of online ads and personalized content), this system will be fair to whoever is being persuaded (Bowles, 2018, pp. 56). It is therefore possible to insert the concept of the social contract into the AI design. The veil of ignorance could also be used, for example, in autonomous vehicles; the system would not know who is who (so-called contractualism).

In this context, for example, Rawls’s theory of justice is used in practice in Singapore to gain the confidence of citizens in technology and strengthen its digital democracy (Keen, 2019). The use of the concept of the social contract can also be found in Estonia, where it is used for mutual transparency between citizens and the state. The fact that political rules in Estonia have evolved in parallel with electronic democracy has pushed Estonia to the forefront, and it can introduce

elements of not ethics, but actual political philosophy, into the system. The social contract in Estonia works on the prin- ciple of mutual control between citizens and the state: when the state wants to explore any data of a selected citizen, the citizen is immediately notified (Keen, 2019).

However, there is still the question of how to apply this concept between the polarity of technology regulation vs. freedom of the market environment where technologies are evolving. Floridi and the neologism of metatechnology (a ra- tional system of protocols, rules and standards for using tech- nology) provide some perspective in this context (Floridi, 2014, pp. 206–215). Floridi offers a solution that could stand up to liberal political theory and the debate between freedom (including innovation) and regulation (protection of the envi- ronment and society). On the one hand, in the application, introduction and dissemination of new technologies (algo- rithms, AI), there is no telos (teleologism) that would assign technologies a certain ethical framework that would define their function in society (e.g. the Enlightenment emancipation framework of people as beings liberated from labour or envi- ronmental protection – which are countervalues to the free market and progress in general). Today, technology, like all commodities, has the main goal of accumulating capital, which can be problematic because it does not take into ac- count the limits of nature and the exploitation of human re- sources (as we have seen above). So Floridi offers a peculiar solution to this situation, in which no one will dictate anyone how to handle technology, but we will create such rules that it is also appropriate to look at the possibilities of technology from the point of view that we will be aware of what we cannot do. Thus, ultimately, it should not be a question of ethics but of rational restructuring of the market environment.

It is argued that information and communication technol- ogies can play an important role in addressing environmen- tal issues, but the negatives of their use are forgotten. One problem for instance is the energy intensity of these systems (Internet of Things, clouds and Big Data). Industry 4.0 can- not go without large data storages and demanding computa- tional performance. Floridi talks about the technological gambit in this context (2014, pp. 212–215). Gambit means (especially in chess) sacrificing something for a future gain. We rely on inventing potential tools to save the environ- ment, but they are only positive if they are used rationally, in specific individual places, while their widespread use is detrimental. Floridi’s metatechnology is the way to control the dangerous nature of the ecological gambit.

Funding The paper was realized in the frame of the projects (Grant TA ČR Éta, no. TL01000299) “Vývoj teoreticko aplikačních rámců pro sociální změnu v realitě transformace průmyslu.” Financial support was received from the Technology Agency of the Czech Republic.

202 Soc (2021) 58:196–203

Declarations

Conflict of Interest The authors declare no competing interests.

References

Allmer, T. (2017). Critical Theory and Social Media: Between Emancipation and Commodification (1st ed.). Routledge.

Bowles, C. (2018). Future Ethics (1st ed.). NowNext Press. Bridle, James. 2019. New Dark Age: Technology and the End of the

Future (Reprint ed.).Verso Chandler, D., Fuchs, Ch.. (2019). Digital Objects, Digital Subjects:

Interdisciplinary Perspectives on Capitalism, Labour and Politics in the Age of Big Data (Illustrated ed.). University of Westminster Press.

Feenberg, A. (2009). Critical Theory of Technology. In: Friis, Jan Kyrre Berg Olsen, Stig Andur Pedersen A, Vincent F. Hendriks (ed.). A companion to the philosophy of technology. Wiley-Blackwell

Feenberg, A. 2014. Democratic rationalization: Technology, power, and freedom. In: Scharff, R. et al. (eds.) Philosophy of Technology - the Technological Condition: An Anthology. Wiley Blackwell.

Floridi, L. (2014]. The Fourth Revolution: How the infosphere is reshaping human reality. Oxford University Press.

Floridi, L., Cowls, J., Beltrametti, M. et al. (2018). AI4People—An Ethical Framework for a Good AI Society: Opportunities, Risks, Principles, and Recommendations. Minds and Machines 28, 689– 707. https://doi.org/10.1007/s11023-018-9482-5

Greenfield, A.. (2017). Radical technologies: the design of everyday life. Verso.

Harari, Y. N. (2018). Homo Deus: A Brief History of Tomorrow (Illustrated ed.). Harper Perennial.

Hawksworth, J., Berriman, R., Goel, S. (2018). Will robots really steal our jobs? An International Analysis of the Potential Long Term Impact of Automation. PricewaterhouseCoopers LLP.

Horkheimer, M.., Adorno, T. (2007). Dialectic of Enlightenment (Cultural Memory in the Present). Stanford University Press.

Jameson, F. (1992). Postmodernism, or, The Cultural Logic of Late Capitalism. Duke University Press.

Jankel, N. S. (2015). AI vs. Human Intelligence: Why Computers Will Never Create Disruptive Innovations, Huffington Post. Available

from: http://www.huffingtonpost.com/nick-seneca-jankel/ai-vs- human-intelligence_b_6741814.html. Accessed 22 April 2021.

Keen, Andrew. 2019. How to Fix the Future. Grove Press. Kowalikova, P., Polak, P., Rakowski, R. (2020). The Challenges of

Defining the Term “Industry 4.0”. Society. 57: 631–636. https:// doi.org/10.1007/s12115-020-00555-7

Makridakis, S. (2017). The forthcoming Artificial Intelligence (AI) rev- olution: Its impact on society and firms. Futures. 90:46–60.

Matochova, J., Kowalikova, P., Rakowski, R.. (2019). Social science's dimension of engineering curriculum innovation. R&E-SOURCE. S17:229-234.

Rawls, J..(1999). A Theory of Justice. Belknap Press: An Imprint of Harvard University Press.

Ross, A. (2017). The Industries of the Future. Simon & Schuster. Veitas, V. and Weinbaum, D. (2017). Living Cognitive Society: A ‘dig-

ital’ World of Views, Technological Forecasting and Social Change 114:16-26. https://www.sciencedirect.com/science/article/ abs/pii/S0040162516300610?via%3Dihub. Accessed 1 April 2021

Publisher’s Note Springer Nature remains neutral with regard to jurisdic- tional claims in published maps and institutional affiliations.

Roman Rakowski is a senior lecturer at the Technical University of Ostrava, Czech Republic. His current research on the development of the theoretical-application frameworks for a social change in the reality of the transformation of industry is funded by a grant from the Technology Agency of the Czech Republic.

Petr Polak has been at the Mendel University in Brno, Czech Republic since 2019; between 2005-2018 he was at the Universities in Australia (Swinburne) and Brunei (UBD), from a post-doctoral up to Associate Professor positions. Before his academic career he worked for 10 years in finance and treasury in various senior positions with multinational corporations, such as Dalkia (now Veolia) and Electrolux.

Petra Kowalikova has been a senior lecturer at the Technical University of Ostrava, the Czech Republic since 2007. She holds a Ph.D. in Sociology from Palacky University, Olomouc, Czech Republic. Her research has focused on the sociology of organizations, the labor market, and corporate culture issues.

203Soc (2021) 58:196–203