CC BY
34
www.hjournal.ru
Journal of Institutional Studies, 2022, 14(1): 108-126 DOI: 10.17835/2076-6297.2022.14.1.108-126
INSTITUTIONAL FRAMEWORK FOR THE DEVELOPMENT OF ARTIFICIAL INTELLIGENCE IN THE INDUSTRY
ANASTASIA Y. NIKITAEVA,
Southern Federal University, Rostov-on-Don, Russia, e-mail: [email protected];
ABDEL-BADEEH M. SALEM,
Ain Shams University, Cairo, Egypt, e-mail: [email protected]
Citation: Nikitaeva A.Y., Salem A.-B.M. (2022). Institutional framework for the development of artificial intelligence in the industry. Journal of Institutional Studies 13(1), 108-126. DOI: 10.17835/20766297.2022.14.1.108-126
The article is devoted to the institutions of dissemination and application of artificial intelligence in industry. Artificial intelligence (AI) is currently one of the most dynamically developing technologies and outcomes of the Fourth Industrial Revolution with a huge transformational impact on the economy. The article further confirms the inclusion of this technology in all industrial frontiers of recent years. In industry, artificial intelligence has a high potential of use with prodigious positive effects, but this potential and positive results are limited by insufficiently designed institutional framework for the development of artificial intelligence. To establish a way of institutionalizing AI in industry, the article systematizes the drivers and limiting factors of its cost-effective deployment in industrial companies. Based on this, the authors outlined a conceptual institutional framework for artificial intelligence in industry, including institutions of different levels as well as formal and informal institutions. The stimulating and limiting function of institutions in the deployment of AI is considered from the strategic perspective and operational regulation. The article substantiates the priority of artificial intelligence legislation, which goes beyond both individual countries and institutional conditions focused on a specific technology. It is necessary to develop the digital economy, activate innovations, create a competitive environment, etc. The authors have confirmed the importance of a broader institutional context of economic and technological development in the context of Industry 4.0. The article also pays attention to industry standards and ethical standards for the dissemination of artificial intelligence. At the same time, the influence of the institute of trust, partnerships, and digital corporate culture on the adoption and deployment of artificial intelligence technologies in industrial companies is taken into account. It is determined that, to understand and accept AI (include it into decision-making processes and business practices), institutions are required to make technologies more understandable for perception.
Keywords: artificial intelligence; institutes; industrial enterprises; industrial development; Industry 4.0; smart manufacturing systems
JEL: D20, L50, O14, O32, O34
© HuKMTaeBa A.£>., CaAeM A.-B.M., 2022
ИНСТИТУЦИОНАЛЬНЫЕ ОСНОВЫ РАЗВИТИЯ ИСКУССТВЕННОГО ИНТЕЛЛЕКТА В ПРОМЫШЛЕННОСТИ
НИКИТАЕВА АНАСТАСИЯ ЮРЬЕВНА,
Южный федеральный университет, Ростов-на-Дону, Россия, email: [email protected];
АБДЕЛ-БАДИХ М. САЛЕМ,
Университет Айн-Шамс, Каир, Египет, e-mail: [email protected]
Цитирование: Никитаева А.Ю., Салем А.-Б.М. (2022). Институциональные основы развития искусственного интеллекта в промышленности. Journal of Institutional Studies 13(1), 108—126. DOI: 10.17835/2076-6297.2022.14.1.108-126
Статья посвящена исследованию институциональных условий распространения и применения искусственного интеллекта в промышленной сфере. Искусственный интеллект в настоящее время относится к числу технологий Четвертой промышленной революции, которые развиваются наиболее динамично и оказывают сильное трансформационное воздействие на экономику. Это подтверждает включение данной технологии во все индустриальные фронтиры последних лет. В промышленности искусственный интеллект потенциально обладает широкими возможностями использования с положительными эффектами. Но эти возможности и эффекты ограничены недостаточно сформированными институциональными рамками разработки и применения искусственного интеллекта. Для определения приоритетных направлений институционали-зации искусственного интеллекта в промышленности в статье систематизированы драйверы и лимитирующие факторы его экономически эффективного развертывания в индустриальных компаниях. На основе этого авторами очерчена концептуальная институциональная рамка развития искусственного интеллекта, включающая институты разных уровней, формальные и неформальные институты. Кроме того, отражена стимулирующая и ограничивающая функция институтов в развитии искусственного интеллекта с позиции стратегической перспективы и оперативного регулирования. В статье обоснована приоритетность нормативно-правового регулирования искусственного интеллекта, выходящего как за рамки отдельных стран, так и за рамки институциональных условий, ориентированных на конкретную технологию. Важность развития цифровой экономики в целом, активизации инновационной деятельности, конкурентной среды и др. позволили сделать вывод о необходимости более широкого институционального контекста экономико-технологического развития в условиях Индустрии 4.0. В статье также уделяется внимание отраслевым стандартам и этическим стандартам распространения искусственного интеллекта. Наряду с этим, учитывается влияние института доверия, партнерских отношений и цифровой корпоративной культуры на принятие и применение технологий искусственного интеллекта в промышленных компаниях. Определено, что для понимания и признания искусственного интеллекта, его включения в процессы принятия решений и бизнес-практики, требуются институты, которые способны сделать соответствующие технологии более понятными для восприятия.
Ключевые слова: институты; искусственный интеллект; промышленные предприятия; развитие промышленности; Индустрия 4.0, умные производственные системы
Introduction
Artificial intelligence (AI) is one of the most dynamically developing and promising digital technologies. IThis is confirmed by the inclusion of artificial intelligence in the top 5 vast majority of technological and industrial frontiers in recent years (2019 Edelman AI Survey, 2019), the leading position of AI in monitoring global trends in digitalization (Мониторинг глобальных трендов цифровизации, 2020), the growing volume of publications, patents and investments in this area (Mutascu, 2021; Мониторинг глобальных трендов цифровизации, 2020), an increasing number of AI applications and types of tasks solved (Zeba et al., 2021; Fujii and Managi 2018). Industry is one of the areas where the use of AI can bring significant positive results due to a variety of application areas. The use of artificial intelligence makes it possible to transfer the industrial segment of the economy to a new technological level. It resuts in bthe increase of economic efficiency of industrial enterprises significantly. Moreover, artificial intelligence can radically transform existing socioeconomic, financial, and industrial ecosystems (Walsh, Levy and Bell, 2019, The New Physics, 2018; Экосистемы в пространстве, 2020; Technology Vision, 2021).
At the same time, the development of artificial intelligence has reached a level where, along with its technological limitations, the opportunities and prospects for using this technology in industry are largely determined by relevant institutional factors. Institutions (rather, their absence or insufficient effectiveness) are one of the main limitation of the wide application and dissemination of AI in the industrial sphere. Current researches in the field of AI and its application in industry are connected more on the technical and IT-side. At the same time, there is a lack of research on the institutionalization of artificial intelligence in industry. In modern scientific literature, there is a gap in theoretical research on the institutional framework that performs an activating and limiting role in the development of artificial intelligence. With this in mind, this research aims to explore the content, structure, and priorities of the establishment of an institutional framework for the development of artificial intelligence in the industrial sphere. To achieve this objective scientometric analysis was first carried out by keywords (Artificial intelligence, AI development, Institutions, etc.) on the databases of publications included in the ScienceDirect and Scopus. From the selected set of publications, articles related to the social sciences or having an interdisciplinary nature were chosen for further study. The repeated search was carried out, taking into account the references of publications selected at the first stage. This set of publications was supplemented with analytical materials from leading consulting companies operating in the area of digital transformation of the economy, the development of Industry 4.0, and artificial intelligence itself. Further, content analysis and mind-mapping were applied to determine the content of the institutional framework for the development of artificial intelligence, its structure, and priorities for its development.
AI as one of the most promising technologies for Industry 4.0 development
"Artificial intelligence is a complex of technological solutions that allows simulating human cognitive functions (including self-learning and finding solutions without a predetermined algorithm) and obtaining results comparable, at least, with the results of human intellectual activity when performing specific tasks"1. The explosive development of artificial intelligence in recent years is inseparably connected with Industry 4.0. (Schwab, 2016), and the formation of the digital economy. The term "Industry 4.0" has become the core representation of the Fourth Industrial Revolution in the scientific sphere. It is connected with the integration of the physical and digital world and significantly changing socio-economic systems. Accordingly, AI research needs to take into account the context of the overall digital transformation of the economy. AI is a driver, a cross-cutting technology, and a resultant indicator of the formation of the digital economy as a result of the Fourth Industrial Revolution.
In recent years, AI has been ranked first among global digitalization trends according to Rostelecom monitoring data (taking into account data on publications, patents, and investments in digital technologies) (Monitoring of global trends, 2020) (Figure 1). AI is also the most significant cross-cutting
1 См.: Национальная стратегия развития искусственного интеллекта на период до 2030 года. Утверждена Указом Президента Российской Федерации от 10 октября 2019 г. № 490. (https://www.garant.ru/products/ipo/prime/doc/72738946/#1000)
trend in the digitalization of the economy (the trend is most closely related to, and strongly affects, other trends, and is an "umbrella" for other trends, and a basic one for any industry) (Figure 2).
Fig. 1. Rating of global digitalization trends (Мониторинг глобальных трендов, 2020)
However, the degree of intelligence of modern artificial intelligence, its types, and features, the relations of artificial intelligence and human intelligence (including from philosophical and psychological positions) is the subject of many scientific discussions (van der Maas, Snoek and Stevenson, 2021; Neubauer, 2021).
As one of the key priority technologies of industrial development, AI is included in the Emerging Technologies Radar 2021 Gartner (Nguyen, 2020.) and Top Strategic Technology Trends for 2021 Gartner (Top Strategic Technology, 2021). Furthermore it is has been added to the main technological trends of Deloitte's digital frontier (Tech Trends 2021), priority institutionalized technologies of the NTI 2035 of the Russian Federation (National Technology Initiative, 2021), and the Table of disruptive technologies of Imperial College London (Table of disruptive, 2018). AI is one of the key technological trends of Industry 4.0 for the past ten years of its implementation (Ghobakhloo, Fathi and Iranmanesh, 2021) and ranks first among the five biggest technological trends (2021), according to Forbes (Marr,
2020.). According to Accenture research, AI is one of the technologies recognized as catalysts of change, opening up new opportunities for enterprises and allowing them to rethink entire industries (Technology Vision, 2021). According to PWC experts, AI can bring up to $ 15.7 trillion to the global economy by 2030. Of these, $ 6.6 trillion is expected to be obtained due to increased productivity, and $ 9.1 trillion is due to effects on the consumption side (Sizing the prize, PWC). However, AI's contribution to growth will not be uniform. The contribution of AI to economic growth by 2030 may be three or more times higher than over the next five years because of the S-curve pattern of AI adoption. A relatively slow start is predicted due to the high costs and investments required for the deployment of AI. Furthermore, the cumulative effect of competition and an improvement in complementary capabilities is expected to accelerate the AI development (Bughin, Seong and Manyika, 2018).
Fig. 2. Rating of cross-cutting digitalization trends (Рейтинг сквозных трендов, 2020)
Thus, artificial intelligence the biggest technological prospect and opportunity in the digital economy. Along with the recognition of the potential and positive effects of the development of AI technologies, comes the understanding, firstly, of a whole set of risks and challenges accompanying the spread of AI (Gruetzemacher, Dorner and Bernaola-Alvarez, 2021; Mutascu, 2021; Ransbotham Khodabandeh and Fehling, 2019, 2019 Edelman AI Survey, 2019). Secondly, the fact that investing in AI does not mean the simultaneous and widespread use of AI technologies and the immediate achievement of corresponding economic effects (Davenport, Zhang 2021, Ransbotham et al., 2019), and thirdly, that it requires a set of institutions to ensure the adequate development of AI (Ghulam, 2021). The last point is connected with the recognition of AI's going beyond the limits of the technological space itself and its transformational impact on the economy and society as a whole. Accordingly, the institutionalization of AI should cover both the general conditions for the deployment of intelligent digital solutions in industry and regulate specific behavioral practices of economic entities. In addition, given the large-scale nature of the projected digital transformations, institutions are required to regulate the development and application of AI from a strategic and operational perspective. A forward-looking vision plays a significant role in the context of reducing uncertainty and improving the efficiency of managing the development of artificial intelligence (Gruetzemacher et al., 2021). A holistic approach to forecasting and planning the development of AI is required (Gruetzemacher, 2019). It is prudent to have a clear understanding of the directions of strategic investments and a roadmap for the development of AI (Lee, Davari and Singh, 2018).
It is advisable to consider the institutional framework for artificial intelligence in conjunction with the fields of application, potential effects, and risks of using AI in a specific area of economic activity. This is determined by the significant features of the implementation of AI in various fields (whether it is industry, public administration, medicine, or education). In addition, the actions and impact of institutions differ depending on the sphere of economic activity, its specific, spatial and temporal characteristics (Вольчик, Бережной, 2009). Institutions are explored in this research in the methodological traditions of neo-institutionalism. Accordingly, institutions are understood under D. North's definition of rules, mechanisms that ensure their implementation, and norms of behavior
that structure repetitive interactions between people (North, 1989). The research task of determining the institutional framework for the development of AI is implemented on the example of industry.
At the same time, both formal and informal institutions are considered. Consequently, the structure of institutions and their types are taken into account under D. North's neo-institutional ideas (Popov and Sukharev, 2017). One of the arguments in favor of this approach is the fact that efficiency and ethics, development institutions and institutions of trust, technologies, and politics are considered as strategically related concepts of Industry 4.0, primarily in terms of AI (Walsh et al., 2019; A Survey of Artificial, 2017). As mentioned earlier, the development of AI goes far beyond the technological field. It is about people, processes, culture, and strategy — and, ultimately, about the revision of the relationship between man and machine (How to Turn AI into ROI) and the institutional framework for their transformation.
The essence and possibilities of using artificial intelligence in industry
In recent years, AI has reached a level of development where it is able to truly solve problems and effectively create economic benefits (Zhang, Cui and Zhu, 2020). In industry, this is due to its ability to produce high-quality demand-driven products quickly and cost-effectively. It is implemented by:
(a) - reducing excessive environmental pollution (as a result of optimizing emissions), reducing high energy consumption (Li, Su and Wang, 2021), creating more environmentally friendly and sustainable supply networks (Yu, Zhang and Cao, 2021);
(b) - optimization and distribution of supply chains, which leads to an increase in positive financial results (Yu et al., 2021);
(c) - the use of expert systems for simulating, modeling, and solving complex problems that are traditionally solved with the involvement of experts (Zhang, Lu, 2021);
(d) - the use of intelligent robots that can analyze the information they perceive, control their behavior, respond to changes in the environment and effectively perform complex tasks (Zhang, Lu, 2021);
(e) - application of intelligent decision support systems to improve the efficiency of management activities (Zhang, Lu, 2021);
(f) - creation of new and transformation of existing products, AI is used both in the processes of developing new products, and as an integral part of the products themselves (a vivid example is a self -driving cars in the engineering industry (Zhang, Lu, 2021);
(g) - complementing existing production practices with network and intelligent capabilities, which contributes to increased flexibility and accuracy of production and leads to adaptive, eco-friendly, and smart production (Yu et all., 2021);
(h) - support production processes through the use of AI to select the most suitable material, taking into account the method of application, properties, and behavior during operation (Merayo, Rodríguez-Prieto and Camacho, 2019).
(i) - automation of internal and external processes, including interaction with customers and customer service;
(j) - the use of virtual assistants and smart chatbots to increase customer satisfaction and the speed of response to their requests;
(k) - optimization of business processes, including through the intelligent analysis of sensor data using the Industrial Internet of Things
(l) - the use of AI for collecting and analyzing marketing information (as an example, the analysis of social networks using AI by Coca-Cola for setting up product lines), the development and implementation of promotion programs and interaction with consumers;
(m) - development and adoption of effective predictive solutions (Global AI Adoption Index, 2021) (n)- increasing the customization of products and simplifying their production through automation and the use of intelligent robots (Ribeiro, Lima and Eckhardt, 2021);
(o)- the use of digital doubles for effective modeling of integrated smart manufacturing systems (cyber-physical systems), covering all types of resources, production and management processes, behavioral aspects of the interaction of elements (Leng, Wang and Shen, 2021);
(p) - Using AI to protect data, improve information security and reduce the risks of cyber attacks (Alhayani, Mohammed and Chaloob, 2021);
(q) - Product lifecycle management and reducing of time-to-market (How AI Benefits, 2021).
Thus, AI in the industry can be used in the internal value chain of companies, in external interactions to improve the user experience, in products, and methods of their delivery, digital enhancement of products through services (for example, through innovative business models of servitization) (Ransbotham et al., 2019).
In the AI ecosystem in the industry, the following types of technologies are distinguished that allow implementing the listed areas (Lee et al., 2018):
(a) - data processing and transmission technologies-implement connection and communication functions (covering interaction between production resources in physical space, data transfer and storage from equipment to the cloud, communications between physical and cyberspace);
(b) - analytical technologies that implement the conversion of data into useful information (including the identification of hidden patterns and unknown correlations in production systems);
(c)- platform technologies, including hardware architecture for storage, analysis and feedback of production data; and
(d) - operational technologies refer to a number of decisions made, and actions taken based on information extracted from data.
In addition, AI technological solutions can be built following the stages of activity at industrial enterprises: enterprise resource planning, production management, control, and management of manufacturing processes (Yang et al., 2021). In this case, different solutions will be suitable for discrete and process productions.
Moreover, most AI success stories are focused on improving existing business processes (in sales, marketing, pricing, service, forecasting, production, etc.). However, such partial solutions give relatively small effects. AI allows you to reinvent and rethink many of the processes of industrial companies. So, the true potential is manifested not in doing the same thing better, faster and cheaper, but in doing completely new things altogether. Then we can talk about disruptive digital innovations in the industry and reaching a new level of development (Ransbotham et al., 2019)
At the same time, some circumstances should be noted separately. According to industry leaders, the potential of AI is especially significant in an environment where non-cost factors dominate (Ransbotham et al., 2019, p. 8). In addition, the success of AI application scenarios closely corresponds to the maturity of the industries themselves. Artificial Intelligence is the most cost-effective when combined with other digital technologies (including the Industrial Internet of Things, mobile networks, etc.) (Ribeiro et al., 2021). This justifies the establishment of an institutional framework both for AI and other digital technologies of Industry 4.0.
Success factors and limitations of the use of AI in industry
Institutional conditions should become, according to the authors, the means to removing restrictions and implementing the drivers of AI deployment in industry. In this regard, it is necessary to objectively consider such limitations and drivers.
According to experts, industrial companies implementing AI initiatives should strive for a level of economic return higher than they have achieved so far. Several studies indicate a low economic return on AI, partly because many AI systems have never been deployed. The economic impact is mainly considered either in terms of reducing costs or in terms of generating income.
An IBM 2021 survey found that only 21% of 5,501 companies said they had deployed AI in business, while the rest are studying AI, working on confirming certain concepts, or using ready-made AI applications (Davenport and Zhang, 2021). Similarly, an analysis by VentureBeat shows that 87% of artificial intelligence models are never put into production (Davenport and Zhang, 2021). In 2019, a study by the MIT Sloan Management Review/Boston Consulting Group found that 7 out of 10 companies reported that their investments in AI did not bring any benefit. The gap in AI maturity is growing. While many companies are experiencing difficulties, a significant minority realizes the value of AI and continues to invest more (Ransbotham et al., 2019). It makes sense: if there is no deployment ofproduction, there is no economic value.
Figure 3 shows the results of AI adoption rates around the world (according to the research commissioned by IBM in partnership with Morning Consult).
Figure 4 shows how AI users were divided into segments (in a study conducted by MIT Sloan Management Review and Boston Consulting Group in 2019, 2,555 respondents from 29 industries and 97 countries participated).
Fig. 3. AI adoption rates around the world (Global AI Adoption Index 2021)
Adoption
Pioneers are organizations that both understand and accept artificial intelligence Researchers: Organizations that demonstrate knowledge about AI technologies and applications, but do not deploy beyond the pilot stage
Experimenters: Organizations that pilot or implement AI without a deep understanding. These organizations learn by doing Passive organizations that do not use AI and are poorly versed in technology
Fig. 4. Four Distinct Segments of AI Users (How to Turn, 2019)
We identify and classify the factors limiting the development of AI in the industry on the level of economic entities:
(a) - preferential concentration on the technical side of the development of AI and a lack of consideration of organizational and institutional factors, the narrow focus on the development of technology, insufficient consideration of the overall strategic aspects of digital transformation and changes in the business model (Ransbotham et al., 2019);
(b) - difficulty in scaling and the lagged impact from investments in AI affect the perception of its economic value from the management of companies and users of AI;
(c) - insufficient calibration and configuration of ready-made vendor solutions, which have become very numerous on the market, for the needs and business models of specific companies (Ransbotham et al., 2019);
(d) - insufficient level of expertise and knowledge in the field of AI (AI Adoption Index, 2021);
(e) - increasing complexity of data and data warehouses;
(f) - lack of tools or platforms for developing AI models (Global AI Adoption Index, 2021);
(g) - lack of skills or training to develop and manage reliable AI;
(h) - AI management and management tools that do not work in all data processing environments;
(i) - lack of regulatory guidance from governments or industry;
(j) - the lack of company guidelines for the development of reliable, ethical AI, the lack of a clear policy in this area of the company's activities;
(k) - insufficient interaction of all parties involved in the development, development, and application of AI;
(l) - building models based on data that have an innate bias (social, economic, etc.) (Global AI Adoption Index, 2021);
(m) - insufficiently predictable effectiveness of the application;
(n) - limited ability to explain decisions made using AI (Global AI Adoption Index, 2021, page 8).
In turn, some companies have already achieved an economic return on their investments in AI. Strategic factors of success in finding value from the use of AI include establishing close relationships between the data group and interested business units, choosing projects with tangible value and a clear path to production, building trust from key stakeholders, creating reusable AI products, and creating a management pipeline or funnel leading projects to manufacturing. (Davenport and Zhang, 2021).
In aggregate form, the factors for the effective use of AI in the industry and the corresponding drivers include:
• the alignment of the AI development strategy with the overall organizational development strategy, the incorporation of artificial intelligence initiatives into bigger business transformation efforts, the mutual use of AI and business strategies to support each other;
• proactive risk management and larger, riskier investments in AI as part of an overall well-thought-out strategy, where revenue growth is put above cost reduction;
• coordination of organizational behavior and strategy for creating business value with the help of AI, considering AI as a strategic initiative that requires new behavior, and not just a new technological opportunity;
• -cross-functional use of AI, integration of stakeholders using these technologies, the use of AI in all areas from the internal value chain to interaction with customers and the development of new products and ways to deliver them;
• seniormanagement's confidence in the promising value of AI;
• balance of the development of AI not only in the production but also in the consumer part, which implies the establishment of a favorable environment in which manufacturers can develop and implement AI solutions in close cooperation with businesses. Ensuring that investments in AI production match investments in AI consumption.
• cultivation of non-technical leaders and business users of AI;
• investing in corporate AI development groups, limited import of external specialists, improving the skills of employees in the field of AI, in data and changing processes, along with investments in artificial intelligence technologies themselves, implementing a policy of growing and creating AI talent, rather than external search (Ransbotham et al., 2019);
• good working relationships between the data team and the business units - and a clear focus on tangible value (Davenport and Zhang, 2021);
• building a sustainable partnership with internal and external stakeholders (Business Readiness for, 2021). This requires institutions of interaction within and outside the organization;
• selection of AI projects for investment based on the same return-on-investment criteria as for any other projects, but taking into account the specifics of the technology. When solving any AI problem, it is necessary to set clear expectations and results for both the AI system and the people who interact with it (Business Readiness for, 2021).
Such external factors to economic entities as the pressures of the competition (Sizing the prize, 2017), the COVID-19 pandemic (IBM Global AI Adoption Index, 2021) have become powerful drivers of the deployment of AI technology in the industry.
Establishment of an institutional framework for the development of artificial intelligence in industry
Thus, for the effective development of AI in the industry, a whole set of institutional conditions is required. Conceptually, this complex can be structured as an institutional framework that:
• has a supranational, national, regional, local level;
• includes formal and informal institutions;
• includes institutions that implement stimulating and limiting functions;
• includes both institutions that directly influence the formation and development of AI and broader institutions that determine the conditions for the implementation of Industry 4.0, the digital transformation of the economy and industry as part of it, the innovation activities, and doing business in general;
• the presence of a strategic perspective and operational regulation of the development of AI.
The authors consider in more detail the institutional framework for the development of AI in the
industry.
Currently, the regulatory framework for the development and application of AI (as a significant part of formal institutions) is only in the process of creation. Given the global nature of digital transformations in line with Industry 4.0 and the crucial role of AI in their implementation, internationally coordinated political actions need to ensure the authority and legitimacy of the emerging set of laws regulating AI. Policy and legal initiatives should take into account the domestic and international regulatory framework (to avoid conflicts due to fragmentation and maximize efficiency). From an economic point of view, regulatory coordination will ensure that AI improves socioeconimc welfare rather than exacerbating existing global economic inequality. In addition, supranational coordination will provide broader social and political support for the AI regulatory framework (Walsh et al., 2019; Korinek and Stiglitz, 2019).
At the same time, the regulation of AI takes place within specific countries and goes beyond the direct impact. A favorable institutional environment leads both to the successful development and improved use of technologies (because of optimizing the scale of production, increasing profitability, and productivity) (Ghulam, 2021; Agostino, Tommaso and Nifo, 2020). Empirical results confirm long-term productivity growth in the industry due to technological progress after the development of formal institutions through broader reforms (Ghulam, 2021). Both the creation and implementation of AI technologies are associated with innovation and entrepreneurial activity. Moreover, taking into account the peculiarities of the technology itself, we can talk about the formation of new markets and industrial segments, respectively, not only incremental but also disruptive innovations. As new AI technology requires significant investments, and therefore, reducing information asymmetry and risk predictability. That is why it is necessary to create generally favorable conditions for entrepreneurial and innovative activities. Adding that competitive pressure is one of the drivers of AI development, there is a need for a sufficient level of competition and effective antimonopoly regulation. AI does not develop in isolation, it is successful in collaboration with other technologies of the digital economy. This means that appropriate institutions are required for their balanced development. High digital
maturity can accelerate AI adoption and absorption (Bughin et al., 2018). When using international best practices, it isprudent to implement not the fragmentary but complex import of institutions (Lukashov, Lukashova and Latov, 2021) and an understanding of their interrelations and interfaces.
On the one hand, almost nowhere has the creation and application of AI been sufficiently and fully developed by legislation to date. The problem of limited legal and technical regulation of the basics, conditions, and features of the development, launch, application, integration into other systems of artificial intelligence technologies is typical in most countries (Hohkhh, PegtKHHa, 2018). On the other hand, every branch of law contains elements of digital law. We cannot talk about a vacuum of formal institutes but rather about significant gaps in this area.
Laptev V. A. considers the development of artificial intelligence, related robotics, and legislation defining the legal personality and legal responsibility of artificial intelligence in the context of certain stages. Soon, a robot with artificial intelligence will be an object of the law. An operator or another person who sets its parameters and controls its behavior will be responsible for the actions of artificial intelligence. In the medium term, AI robots acquire legal personalities and act as participants in relations. In the long run, legal personnel will exist for artificial intelligence already in the virtual (digital) space in isolation from the material world. Cyber-physical legal responsibility will have a regulatory and protective function (Laptev, 2019).
Both the population and technical managers of organizations believe that legislative regulation is required for the successful and safe development of AI (Fig. 5).
* PErcentages dD not always add to 100% duE to rounding.
Fig. 5. Opinion about the role of legislation for AI development (2019 Edelman AI SURVEY, 2019: 28)
Along with the regulatory framework of the legal environment, which is essential for everyday practice, it is necessary to consider the strategic perspective of the development of AI technologies in the institutional context. Historically, most studies on technological development and related technological strategies have been concentrated on developed countries with a stable institutional environment (Amankwah-Amoah et al., 2021). While within the framework of Industry 4.0, new technological opportunities are opening up for developing economies. However, unlike developed countries — characterized by a stable legal system and financial institutions, developing economies are often characterized by institutional voids (Amankwah-Amoah et al., 2021) and institutional traps (Volchik, 2019). These voids and hurdles include inefficient sustainable institutions, such as bureaucracy, administrative delays, inadequate disclosure regime, corruption and political instability (Acquaah, 2007), and the unregulated nature of certain areas of the digital economy. The field of application of AI is just one such example.
A window of opportunity for developing countries (for a breakthrough in economic development, the transition from the old to the latest technologies, bypassing the intermediate stages) opens with paradigm changes in technologies and significant changes in institutions (Amankwah-Amoah et al., 2021, Binz, Truffer and Li, 2012). At the same time, the general policy of technological development is also associated with modernization and transformation of the industry structure, especially for developing economies. For example, we can cite the vector of transition from high-speed economic growth to high-quality growth of the Chinese economy (from high-speed growth to high-quality growth). In the context of China, this is a transition from a labor-intensive production chain with low added value to a technological production chain with high added value and continuous improvement of production efficiency. To modernize the industrial structure, the possibilities such as environmental legislation, are used. Based on the data of China, researchers have proved that environmental standards can have a significant positive impact on the modernization of the sectoral structure of the economy, but in conjunction with other development factors and with the effective interaction of the federal center and the regions (Song, Zhang and Zhang, 2021). Another significant part that actualizes the regional level of the framework is related to the need for an infrastructure for the digital economy. Such infrastructure varies significantly in different regions of developing countries (which is confirmed by the experience of, for example, Russia and China).
At the same time, we need to consider the limiting function of institutions. This is determined both by the large-scale transformational potential of AI and the ambiguous nature of its impact on the economy. For example, the dissemination of AI can lead to an increase in unemployment (Mutascu, 2021). At the same time, the study conducted for the most technologically and economically developed countries shows a non-linear relationship between artificial intelligence and unemployment due to the level of inflation. At low levels of inflation, artificial intelligence improves employment. While otherwise, its effect seems to be zero. In this case, when inflation is low, the intensive use of artificial intelligence reduces unemployment as long as the trend towards higher wages is offset by growth and the creation of new jobs (Mutascu, 2021).
It is also important to note the institutions of information security, which, even though formal (including relevant doctrines and regulatory acts), are directly linked to informal institutions, for example, behavioral practices and trust in AI.
Some studies show the role of AI in improving the institutional environment. Justifying that AI can minimize causes of transaction costs (such as uncertainty, information asymmetry, and opportunistic behavior), Shkodinsky S. V. and Nadyseva D. M. argue the validity of the position that artificial intelligence itself is a new economic institution. (Shkodinsky and Nadyseva, 2020).
At the same time, external conditions have an impact on both formal and informal institutions. An example is the COVID-19 pandemic. In the new crisis era, society requires policies that not only eliminate digital isolation but also ensure equality in access to opportunities for individuals and markets for businesses (Amankwah-Amoah et al., 2020).
Ensuring a safe, accurate, effective AI depends on effective management and a flexible regulatory system that encourages innovation. It is also necessary to strengthen public confidence that the goods and services produced by AI meet or exceed basic standards and to preserve the values that society strives for (Walsh et al., 2019). This requires the development of policies, strategies, and national AI development programs. Canada was the first country in the world to adopt a national AI strategy, which has resulted in a fairly wide range of strategic documentation (Dillet, 2018). Germany has announced €3 billion over six years for its AI 'Made in Germany' digital strategy with the aim to boost the country's AI capabilities (Walsh et al., 2019). In 2019 the AI strategy development plan was adopted in National Strategy for the Development of Artificial Intelligence2. An important part of it speaks to the formation of a comprehensive system for regulating public relations arising in connection with the development and use of artificial intelligence technologies.
With the arguments given earlier, it is worth considering strategies for the development of AI as well as strategies and programs for digital transformation, Industry 4.0, and information security.
2 См.: Национальная стратегия развития искусственного интеллекта на период до 2030 года. Утверждена Указом Президента Российской Федерации от 10 октября 2019 г. № 490. (https://www.garant.ru/products/ipo/prime/doc/72738946/#1000)
The development of artificial intelligence leads both to the emergence of new institutions and the transformation of existing institutions. For example, the adoption of the strategy for the development of artificial intelligence in Russia led, firstly, to the creation of a Roadmap for the development of "cutting-edge" digital technology "Neurotechnologies and Artificial Intelligence"3, the signing by the country's largest companies of the first code of ethics of artificial intelligence on October 26, 20214. On April 24, 2020, Federal Law No. 123-FZ "On Conducting an Experiment to Establish Special Regulation to Create the necessary conditions for the Development and Implementation of Artificial Intelligence Technologies in the Subject of the Russian Federation — the Federal City of Moscow and Amendments to Articles 6 and 10 of the Federal Law "On Personal Data" was adopted (ФЗ №123-ФЗ)5. Secondly, the composition of the current national program "Digital Economy of the Russian Federation"6, which included six federal projects, has been supplemented since 2020 by the seventh federal project "Artificial Intelligence". In addition, a new grant competition was launched in 2021 to support projects of small enterprises in the field of AI in the structure of the existing development institute — the Innovation Assistance Fund7. The institution of e-government is being transformed due to the transition to smart government. Higher education is changing through the prioritization of bachelor's and master's degree programs in the field of AI. And this is only what concerns the development and transformation of formal institutions.
It is important to note that most AI strategies are focused particularly on industry. As an example, the Strategic Plan "Made in China, 2025" is aimed at integrating information and production technologies, increasing the variety and volume of intelligent production equipment and products, developing and implementing smart manufacturing strategies, as well as increasing the intelligence of production processes (Yu et al., 2021). As a result of the implementation of this plan, industrial companies are forced to carry out an intellectual transformation to modernize their traditional practices. This has already led to the emergence of a large number of various kinds of intelligent technologies and smart factories in industry leaders, including Huawei, Haier, Gree and Geely, Baosteel, Hikvision (Yu et al., 2021).
Samar Fatima, Kevin C. Desouza, Gregory S. Dawson noted in their study that in 2020, the manufacturing sector was highlighted in 14 country strategic plans for the development of artificial intelligence, including, for example, New Zealand and Finland (Fatima, Desouza and Dawson, 2020). The institutionalization of AI in various strategic plans is made possible through the formation of a regulatory environment, the development of specialized education, the promotion of innovation, management mechanisms and institutions for the protection of intellectual property, and solving issues in the field of ethics and trust in AI (Fatima, Desouza and Dawson, 2020).
Among the institutions that ensure the ethical and humanistic creation and use of AI, industry standards of AI and existing initiatives in this field should be noted (Rana El Kaliouby, 2019). The Association of Technological and Scientific Circles is an initiative of MIT and Harvard on the Ethics and anagement of Artificial Intelligence; an AI Partnership created as a collaboration between large technology companies (including Microsoft, Amazon, Google, Facebook, and Apple). In the future, the combination of ethics and strategy will result in the formation of AI development roadmaps (for example, the AI Roadmap Institute https://www.roadmapinstitute.org/roadmap-comparison).
Ultimately, to be accepted by users - both internal and external - AI systems must be clearly understood. This means that the decision-making framework of AI should be well explained and
3 См.: Дорожная карта развития «сквозной» цифровой технологии «Нейротехнологии и искусственный интеллект». (2019). Министерство цифрового развития, связи и массовых коммуникаций Российской Федерации, Москва. (https://digital.gov.ru/uploaded/files/07102019ii.pdf)
4 См.: Кодекс этики в сфере искусственного интеллекта. 2021. (http://cdn.tass.ru/data/files/ru/kodeks-etiki-ii.pdf)
5 См.: Федеральный закон от 24 апреля 2020 г. N 123-ФЗ «О проведении эксперимента по установлению специального регулирования в целях создания необходимых условий для разработки и внедрения технологий искусственного интеллекта в субъекте Российской Федерации - городе федерального значения Москве и внесении изменений в статьи 6 и 10 Федерального закона «О персональных данных» (https://base.garant.ru/73945195/#ixzz7DmB7yM9C)
6 См.: Программа Цифровая экономика Российской Федерации. Утверждена распоряжением Правительства Российской Федерации от 28 июля 2017 г. № 1632-р. http://static.government.ru/media/files/9gFM4FHj4PsB79I5v7yLVuPgu4bvR7M0.pdf
7 См.: В России начали выдавать гранты на коммерциализацию технологий искусственного интеллекта. 2021. Министерство экономического развития Российской Федерации. (https://www.economy.gov.ru/material/news/v_rossii_nachali_vydavat_granty_na_ kommercializaciyu_tehnologiy_iskusstvennogo_intellekta.html).
determined. It is necessary to protect AI from constantly evolving and emerging new threats. New management and control methods focused on dynamic AI learning processes can help mitigate risks and strengthen confidence in AI (Cobey and Boillet, 2021)
Awareness, understanding of strategic goals, principles, methods, and results of AI are required to adopt it at the informal level. So, institutionalization should cover the entire cycle from development to application/management of AI and its control. Institutionalization can include AI ethics councils, corporate AI development standards, comprehensive AI impact assessment, validation tools, training sessions for management and employees, independent audits (Cobey and Boillet, 2021).
It is also necessary to consider the interconnection of technological and institutional innovations (Сухарев, 2009). On the one hand, institutional conditions are required for technological breakthroughs. On the other hand, the higher the level of technological development, the greater the likelihood of institutional breakthroughs. But it is also necessary to take into account the mutual conditionality of institutional and technological development. Technologically driven changes and their consequences are a complex mix of many socio-technological variables (Woodhead, Stephenson and Morrey, 2018). When Henry Ford began mass production of the Ford Model T car using the technological innovation of the first conveyor, the need for appropriate new institutions (including traffic police, the transformation of road regulation, etc.) began to manifest themselves (Woodhead, 2012).
At the same time, trust in AI systems is closely related to social trust (Nestik, 2019). Low institutional trust increases the techno-humanitarian imbalance when the introduction of new technologies outstrips the ability of society to agree on the rules for their use (Nestik, 2019). Moreover, according to psychological research, trust in cyber-physical systems differs from trust in people and is easily replaced by absolute distrust (Nestik, 2019).
At the organizational level, the impact of informal institutions is usually stronger than formal ones (in the sphere of influence on entrepreneurial activity) (Becker and Woessmann, 2009). This proves that it is necessary to create a culture favorable for digital innovations (both in terms of development and application) and appropriate transition strategies.
In this case, we should consider the routine of new behavior associated with the creation and use of artificial intelligence from the standpoint of evolutionary economics (Nelson et al., 2018). This requires the development of special strategies, the introduction of innovations, training. It will allow not only to implement evolutionary changes quickly enough but also to transfer new routines, practices, and technologies to new generations of economic agents. It is advisable to consider such evolution from the standpoint of an expanded evolutionary synthesis, comprehensively disclosed and developed by Frolov D.P. (Фролов, 2020). This is important because it is a digital culture that is the driver of the digital transformations of any organization that determine new practices for the use of AI (Digital Culture, 2021).
Primarily, we should broadly talk about the comprehension and acceptance the values of digital culture. The most important values are considered below (Westerman, Soule and Eswaran 2021):
• impact-radical change of socio-economic systems through constant innovation;
• speed — fast movement and action until you get all the answers to your questions;
• openness — the use of different information sources and openness as opposed to the concealing of knowledge and information;
• autonomy - allows people with a high degree of discretion to do what is needed, instead of relying on formally structured coordination and policy.
On this value basis, the adoption or improvement of a set of practices based on digital technologies is further carried out. They will determine the actions of employees and the effectiveness of the organization. Culture is more difficult to change than strategy because much of it is unconscious. Institutional support can enforce rapid innovative and experimental actions (for example, within the framework of design thinking), data-based management and decision-making, self-organization-closely related to partner interactions and collaborations (Westerman, Soule and Eswaran, 2021).
The institutionalization of various forms of partnership interactions plays an important role both in the context of the activation of innovation activities (Nikitaeva, 2017) and in terms of the faster spread
of AI technologies as a result of the exchange of knowledge and technologies, the sharing of technologies in partner structures and digital ecosystems. Evolutionary economics and extended evolutionary synthesis can also serve as a theoretical basis for research in this direction (Фролов, 2020).
When implementing disruptive technologies, companies, on the one hand, learn by trial and error. On the other hand, the effectiveness of this process depends on previously accumulated knowledge and technological experience (Paiola et al., 2021). Therefore, institutions supporting experiments, creativity, innovation, interaction, and exchange of experience with partners, clients, and other stakeholders are required to gain access to the necessary knowledge and accumulated experience, including through networks and industrial ecosystems.
The removal of both technological and institutional constraints on the development of AI can be achieved through the coordination and adoption of comprehensive digital production standards based on different conceptual components of industry 4.0 (for example, Zero-Defects Manufacturing, production with zero defects). An exploration of such standards (and their interconnection) for creating digital production platforms was conducted by (Nazarenko, Sarraipa and Camarinha-Matos, 2021). Researches take into account standards' compliance at different level (assets, integration, communication, information, functionality, business), hierarchies (ranging from product to enterprise, and inter-company connected production chains), and stages of the life cycle and value creation of a production enterprise
In general, we can conclude that to understand and recognize AI, its inclusion in decision-making processes and business practices, institutions are required to make the relevant technologies more understandable, concrete, and tangible for perception. One example of this is the introduction of innovative theaters, in which, with the help of props and interactive controls, users not only receive information, but also emotional attachment to AI and an impression of its capabilities (Ransbotham et al., 2019). It allows us to overcome resistance to change and create new behavioral practices for creating and using AI.
Conclusions
The establishment of such an institutional framework (covering formal and informal institutes of different levels) will result in the successful development and application of artificial intelligence in industry. At the same time, it is necessary to solve objectives in three paths simultaneously. First, to create new institutions for regulating the development and dissemination of AI. Second, to modernize existing institutions. Third, to consolidate new routines and effective behavioral patterns. In this case, the effects will be positive, covering both improvements for enterprises and an innovative technological transformation of industry (while maintaining control over risks). Moreover, then it allows synchronizing mutual stimulation of technological and institutional innovations. At the same time, further relevant areas of research are related to the study and assessment of the degree of maturity of the institutional environment for the development of artificial intelligence. It is also of scientific interest to identify ways and mechanisms that contribute to the breakthrough, rather than incremental, development of artificial intelligence from technological and institutional positions.
СПИСОК ЛИТЕРАТУРЫ/REFERENCES
Вольчик В.В. (2019). Институциональные ловушки в сфере образования и науки в условиях оптимизации. Журнал экономической теории 16(4): 783-795 [Volchik V.V. (2019). Institutional Traps in the Education and Science Sector under the Conditions of Optimisation. Russian Journal of Economic Theory 16(4): 783-795 (in Russian.)] Вольчик В.В., Бережной В.И. (2009). Иерархии и комплементарность институтов в рамках хозяйственного порядка. TERRA ECONOMICUS 7(2): 65-73 [Volchik V.V., Berezhnoy V.I. (2009). Hierarchies and complementarity of institutions within the economic order. TERRA ECONOMICUS 7(2): 65-73 (in Russian.)] Лаптев В.А. (2019). Понятие искусственного интеллекта и юридическая ответственность за его работу. Право. Журнал Высшей школы экономики (2): 79-102. [Laptev V.A. (2019) Artificial Intelligence and Liability for its Work. Pravo. Zhurnal Vysshey shkoly ekonomiki (2): 79-102 (in Russian.)]
Лукашов Н.В., Лукашова С.С., Латов Ю.В. (2021). Парадоксы российских программ информатизации. Journal of Institutional Studies 13(1): 115-134. DOI: 10.17835/20766297.2021.13.1.115-134 [Lukashov N.V., Lukashova S.S., Latov Y.V. (2021). Paradoxes of the Russian digitalization programs. Journal of Institutional Studies 13(1): 115-134. DOI: 10.17835/2076-6297.2021.13.1.115-134 (in Russian.)] Мониторинг глобальных трендов цифровизации 2020. Ростелеком. (https://www.company.rt.ru/ upload/iblock/6e0/R0STELEmM_TRENDS2020_INTERACTIVE_FINAL.pdf) [Monitoring of global digitalization trends 2020. Rostelecom. (in Russian.)] Нестик Т.А. (2020). Искусственный интеллект в государстве: этика и доверие. Этика и «Цифра»: Этические проблемы цифровых технологий. Аналитический доклад. (https://ethics.cdto. ranepa.ru/3_6#link152) [Nestik T.A. (2020). Artificial intelligence in the state: ethics and trust. Ethics and "Digital": Ethical problems of digital technologies. Analytical report. (in Russian.)] Никитаева А.Ю. (2017). Институциональная структура региона в контексте инновационного развития промышленности. Журнал институциональных исследований 9(1): 134-149. DOI: 10.17835/20766297.2017.9.1.134-149. [Nikitaeva A.Yu. (2017). Institutional structure of the region in the context of innovative development of industry. Journal of Institutional Studies 9(1): 134-149. (in Russian.)] Понкин И.В., Редькина А.И. (2018). Искусственный интеллект с точки зрения права. Вестник Российского университета дружбы народов. Серия: Юридические науки 22(1): 91-109, doi: 10.22363/2313-2337-2018-22-1-91-109 [Ponkin I.V., Redkina A.I. (2018). Artificial Intelligence from the Point of View of Law. RUDN Journal of Law 22(1): 91-109. Available from: doi: 10.22363/2313-2337-2018-22-1-91-109. (in Russian.)] Рейтинг сквозных трендов. (2020). Отчет. Ростелеком. (https://digitaltrends.rt.ru/reports/over_
arching_rating) [Rating of end-to-end trends. (2020). Report. Rostelecom. (in Russian.)] Сухарев О.С. (2009). Информационная экономика: аспекты развития. Финансы и кредит (5): 8-21. [Sukharev O.S. (2009). Information economy: aspects of development. Finance and credit (5): 8-21. (in Russian.)]
Фролов Д.П. (2020). Эволюционная экономика на пике и в кризисе: перспектива новой парадигмы. Journal of Institutional Studies 12(1): 019-037. D0I:10.17835/2076-6297.2020.12.1.019-037. [Frolov D.P. (2020). Evolutionary Economics at the Peak and in Crisis: Prospects for the New Paradigm. Journal of Institutional Studies 12(1): 019-037. (in Russian.)] Шкодинский С.В., Надысева Д.М. (2020). Особенности функционирования искусственного интеллекта в условиях современных институциональных изменений. Креативная экономика 14(10): 2243-2252. doi: 10.18334/ce.14.10.110900 [Shkodinskiy S.V., Nadyseva D.M. (2020). The artificial intelligence particularities in the context of modern institutional changes. Creative Economy 14(10): 2243-2252. (in Russian.)] Экосистемы в пространстве новой экономики: монография (2020). / науч. ред.: М.А. Боровская, Г.Б. Клейнер, Н.Н. Лябах, М.А. Масыч, Л.Г. Матвеева, И.К. Шевченко ; Южный федеральный университет. Ростов-на-Дону; Таганрог: Издательство Южного федерального университета. [Ecosystems in the space of the new economy: monograph (2020). / scientific editor: M.A. Borovskaya, G.B. Kleiner, N.N. Lyabakh, M.A. Masych, L.G. Matveeva, I.K. Shevchenko; South Federal University. Rostov-on-Don; Taganrog: Publishing House of the Southern Federal University. (in Russian.)] 2019 Edelman AI SURVEY (2019), Edelman. (https://www.edelman.com/sites/g/files/aatuss191/
files/2019-03/2019_Edelman_AI_Survey_Whitepaper.pdf) A Survey of Artificial General Intelligence Projects for Ethics, Risk, and Policy. (2017). Global Catastrophic Risk Institute Working Paper 17-1. (Available online: https://papers.ssrn.com/sol3/ papers.cfm?abstract_id=3070741) Acquaah M. (2007). Managerial social capital, strategic orientation, and organizational performance
in an emerging economy. Strategic Manag. J. 28: 1235-1255 Agostino M., Di Tommaso M.R., Nifo A., Rubini L., Trivieri F. (2020). Institutional quality and firms' productivity in European regions. Reg. Stud. : 1-14.
Alhayani B., Mohammed H.J., Chaloob I.Z., Ahmed J.S. (2021). Effectiveness of artificial intelligence techniques against cyber security risks apply of IT industry. Materials Today: Proceedings, https:// doi.org/10.1016/j.matpr.2021.02.531 Amankwah-Amoah J., Debrah Y.A., Yu W., Lin Z., Danso A., Adomako S. (2021). Technology strategies in emerging economies: Emerging issues, challenges and new research agenda. Technological Forecasting and Social Change 170: 120881, https://doi.org/10.1016/j.techfore.2021.120881. Amankwah-Amoah J., Khan Z., Wood G. (2020). COVID-19 and business failures: the paradoxes of
experience, scale, and scope for theory and practice. Eur. Manag. J., 10.1016/j.emj.2020.09.002 Artificial Intelligence. The next frontier in investment management. Deloitte. https://www2. deloitte.com/content/dam/Deloitte/global/Documents/Financial-Services/fsi-artificial-intelligence-investment-mgmt.pdf
Becker S., Woessmann L. (2009). Was Weber wrong? A human capital theory of protestant economic
history. Q.J. Econ. 124(2): 531-596. Business Readiness for AI: What Your Organization Needs to Know. (2021). MIT SMR CONNECTIONS.
https://sloanreview.mit.edu/offer-sas-business-readiness-for-ai/ Cobey C., Boillet J. (2021). How do you teach AI the value of trust? EY. (https://www.ey.com/en_gl/
digital/how-do-you-teach-ai-the-value-of-trust) Fujii H., Managi S. (2018). Trends and priority shifts in artificial intelligence technology invention: A
global patent analysis. Econ. Anal. Policy 58: 60-69. Binz C., Truffer B., Li L., Shi Y., Lu Y. (2012). Conceptualizing leapfrogging with spatially coupled innovation systems: the case of onsite wastewater treatment in China Technol. Forecast. Soc. Change 79:155-171
Bughin J., Seong J., Manyika J., Chui M., Joshi R. (2018). Notes from the AI frontier modeling the impact of ai on the world economy. McKinsey Global Institute https://www.mckinsey.com/ featured-insights/artificial-intelligence/notes-from-the-AI-frontier-modeling-the-impact-of-ai-on-the-world-economy#
Davenport T.H., Zhang R. (2021). Achieving Return on AI Projects. MIT Sloan management review. August. (https://sloanreview.mit.edu/article/achieving-return-on-ai-projects/?utm_ source = newsletter&utm_medium = email&utm_content = good%20working%20 relationships&utm_campaign=Enews%20AI%208/10/2021) Digital Culture: The Driving Force of Digital Transformation. 2021. World Economic Forum Digital Culture Guidebook. (https://www3.weforum.org/docs/WEF_Digital_Culture_Guidebook_2021.pdff) Fatima S., Desouza K.C., Dawson G.S. (2020). National strategic artificial intelligence plans: A multidimensional analysis. Economic Analysis and Policy 67: 178-194, https://doi.org/10.1016Zj.eap.2020.07.008. Ghobakhloo M., Fathi M., Iranmanesh M., Maroufkhani P., Morales M.E. (2021). Industry 4.0 ten years on: A bibliometric and systematic review of concepts, sustainability value drivers, and success determinants. Journal of Cleaner Production 302: 127052, https://doi.org/10.1016/j.jclepro.2021.127052 Ghulam Y. (2021). Institutions and firms' technological changes and productivity growth. Technological
Forecasting and Social Change 171: 120993, https://doi.org/10.1016/j.techfore.2021.120993. Global AI Adoption Index 2021. New research commissioned by IBM in partnership with Morning Consult. https://filecache.mediaroom.com/mr5mr_ibmnews/190846/IBM's%20Global%20AI%20 Adoption%20Index%202021_Executive-Summary.pdf Gruetzemacher R. (2019). A holistic framework for forecasting transformative AI. Big Data Cognitive Comput 3(3): 35 doi: 10.3390/bdcc3030035
Gruetzemacher R., Dorner F.E., Bernaola-Alvarez N., Giattino C., Manheim D. (2021). Forecasting AI progress: A research agenda. Technological Forecasting and Social Change 170: 120909, https:// doi.org/10.1016/j.techfore.2021.120909. Kaliouby R.El. (2019). How do we build trust between humans and AI? https://www.weforum.org/
agenda/2019/08/can-ai-develop-an-empathetic-bond-with-humanity van der Maas H.L.J., Snoek L., Stevenson C.E. (2021). How much intelligence is there in artificial intelligence? A 2020 update. Intelligence 87: 101548, https://doi.org/10.1016/j.intell.2021.101548.
How AI Benefits Business Leaders: Delivering ROI, Improved Time to Market, and Strategic Success. (2021). EBOOK. DataRobot//https://www.datarobot.com/wp-content/uploads/2021/08/ How-AI-Benefits-Business-Leaders-Delivering-ROI-Improved-Time-to-Market-and-Strategic-Success-Executive-Persona-Ebook.pdf?mkt_tok=Nzg5LVVXVi01OTcAAAF-8rbjflL3yd4MQcTzptUuAa5l7i_QGYl_sv2uG5bk23saGpvetaohSM32BqQeIW8oZBW6gPNGB MoxJ4CLaUuCVWHwN64zsgDlMj4TChpji98R
How to Turn AI into ROI. https://www.bcg.com/ru-ru/featured-insights/how-to/roi-of-ai
Korinek A., Stiglitz J.E. (2019). 'Artificial Intelligence and Its Implications for Income Distribution and Unemployment', in Agrawal, A.K., Gans J., and Goldfarb A. (eds) The Economics of Artificial Intelligence: An Agenda. National Bureau of Economic Research, University of Chicago Press. Available at: http://www.nber.org/chapters/c14018.
Lee J., Davari H., Singh J., Pandhare V. (2018). Industrial Artificial Intelligence for industry 4.0-based manufacturing systems. Manufacturing Letters. 18: 20-23, https://doi.org/10.1016/j. mfglet.2018.09.002
Leng J., Wang D., Shen W., Li X., Liu Q., Chen X. (2021). Digital twins-based smart manufacturing system design in Industry 4.0: A review. Journal of Manufacturing Systems 60: 119-137, https:// doi.org/10.1016/j.jmsy.2021.05.011.
Li F., Su Z., Wang G. (2021). An effective integrated control with intelligent optimization for wastewater treatment process. Journal of Industrial Information Integration 24: 100237, https:// doi.org/10.1016/j.jii.2021.100237.
Marr B. (2020). The 5 Biggest Technology Trends In 2021 Everyone Must Get Ready For Now. https://www.forbes.com/sites/bernardmarr/2020/09/14/the-5-biggest-technology-trends-in-2021-everyone-must-get-ready-for-now/?sh=34ae7ec51b82
Merayo D., Rodríguez-Prieto A., Camacho A.M. (2019). Comparative analysis of artificial intelligence techniques for material selection applied to manufacturing in Industry 4.0. Procedía Manufacturing 41: 42-49, https://doi.org/10.1016/j.promfg.2019.07.027.
Mutascu M. (2021). Artificial intelligence and unemployment: New insights. Economic Analysis and Policy 69: 653-667, https://doi.org/10.1016/j.eap.2021.01.012.
Nazarenko A.A., Sarraipa J., Camarinha-Matos L.M., Grunewald C., Dorchain M., Jardim-Goncalves R. (2021). Analysis of relevant standards for industrial systems to support zero defects manufacturing process. Journal of Industrial Information Integration 23: 100214, https://doi. org/10.1016/j.jii.2021.100214.
Nelson R.R., Dosi G., Helfat C.E., Pyka A., Winter S.G., Saviotti P.P., Lee K., Malerba F., Dopfer. K. (2018). Modern Evolutionary Economics: An Overview. Cambridge: Cambridge University Press.
Neubauer A.C. (2021). The future of intelligence research in the coming age of artificial intelligence - With a special consideration of the philosophical movements of trans- and posthumanism, Intelligence 87: 101563, https://doi.org/10.1016/j.intell.2021.101563.
Nguyen T.H. (2020). Gartner Launches Emerging Technologies Radar 2021. https ://blogs.gartner. com/tuong-nguyen/2020/12/07/gartner-launches-emerging-technologies-radar-2021/
North D.C. (1989). Institutions and Economic Growth: An Historical Introduction. World Development 17(9): 1319-1332.
Paiola M., Schiavone F., Khvatova T., Grandinetti R. (2021). Prior knowledge, industry 4.0 and digital servitization. An inductive framework. Technological Forecasting and Social Change 171: 120963, https://doi.org/10.1016/j.techfore.2021.120963.
Popov E.V., Sukharev O.S. (2017). Institutional factors of economic changes: development trajectories of Douglas North's theory. Perm University Herald. Economy 12(3): 338-359.
Ransbotham S., Khodabandeh S., Fehling R., LaFountain B., Kiron D. (2019). "Winning With AI," MIT Sloan Management Review and Boston Consulting Group, October. https://image-src.bcg. com/Images/Final-Final-Report-Winning-With-AI-R_tcm9-231660.pdf
Ribeiro J., Lima R., Eckhardt T., Paiva S. (2021). Robotic Process Automation and Artificial Intelligence in Industry 4.0 — A Literature review. Procedia Computer Science 181: 51—58, https:// doi.org/10.1016/j.procs.2021.01.104.
Schwab K. (2016). The Fourth Industrial Revolution. World Economic Forum.
Sizing the prize. (2017). What's the real value of AI for your business and how can you capitalise? PWC. https://www.pwc.com/gx/en/issues/analytics/assets/pwc-ai-analysis-sizing-the-prize-report. pdf Song Y., Zhang X., Zhang M. (2021). The influence of environmental regulation on
industrial structure upgrading: Based on the strategic interaction behavior of environmental regulation among local governments. Technological Forecasting and Social Change 170: 120930, https://doi.org/10.1016Zj.techfore.2021.120930.
Table of disruptive technologies. (2018). Imperial College London. https://www.imperial.ac.uk/ media/imperial-college/administration-and-support-services/enterprise-office/public/Table-of-Disruptive-Technologies.pdf
Tech Trends 2021. (2021). Deloitte. https://www2.deloitte.com/content/dam/insights/articles/6730_ TT-Landing-page/DI_2021-Tech-Trends.pdf
Technology Vision 2021. (2021). Leaders Wanted. Accenture. https://www.accenture.com/us-en/ insights/technology/_acnmedia/Thought-Leadership-Assets/PDF-3/Accenture-Tech-Vision-2021-Full-Report.pdf
The New Physics of Financial Services. Understanding how artificial intelligence is transforming the financial ecosystem. (2018). World Economic Forum. Prepared in collaboration with Deloitte. http://www3.weforum.org/docs/WEF_New_Physics_of_Financial_Services.pdf
Top Strategic Technology Trends for 2021. (2021). Edited by Brian Burke, Research Vice President, Gartner. https://emtemp.gcom.doud/ngw/globalassets/en/information-technology/documents/ insights/top-tech-trends-ebook-2021.pdf
Walsh T., Levy N., Bell G., Elliott A., Maclaurin J., Mareels I.M.Y., Wood F.M. (2019). The effective and ethical development of artificial intelligence: An opportunity to improve our wellbeing. Report for the Australian Council of Learned Academies, www.acola.org.https://acola.org/wp-content/ uploads/2019/07/hs4_artificial-intelligence-report.pdf
Westerman G., Soule D.L., Eswaran A. (2021) Building Digital-Ready Culture in Traditional Organizations. Special collection. Bringing digital transformation to your organization. MIT SLOAN MANAGEMENT REVIEW.
Woodhead R.M. (2012). What is technology? Int. J. Sociotechnol. Knowl. Dev. 4(2): 1-13. (http://dx.doi. org/10.4018/jskd.2012040101).
Woodhead R., Stephenson P., Morrey D. (2018). Digital construction: From point solutions to IoT ecosystem. Automation in Construction 93: 35-46.
Yang T., Yi X., Lu S., Johansson K.H., Chai T. (2021) Intelligent Manufacturing for the Process Industry Driven by Industrial Artificial Intelligence. Engineering. doi: https://doi.org/10.1016/j. eng.2021.04.023
Yu Y., Zhang J.Z., Cao Y., Kazancoglu Y. (2021). Intelligent transformation of the manufacturing industry for Industry 4.0: Seizing financial benefits from supply chain relationship capital through enterprise green management, Technological Forecasting and Social Change 172: 120999, https:// doi.org/10.1016/j.techfore.2021.120999.
Zeba G., Dabic M., Cicak M., Daim T., Yalcin H. (2021). Technology mining: Artificial intelligence in manufacturing, Technological Forecasting and Social Change 171: 120971, https://doi. org/10.1016/j.techfore.2021.120971.
Zhang C., Lu Y. (2021). Study on artificial intelligence: The state of the art and future prospects. Journal of Industrial Information Integration 23: 100224, https://doi.org/10.1016/jj.jii.2021.100224.
Zhang Z., Cui P., Zhu W. (2020). Deep learning on graphs: a survey, IEEE Trans. Knowl. Data Eng. Early Access.
