1. Introduction: Big Tech Acceleration to Hegemony (2012-2022)
The technology sector underwent a dramatic acceleration over the last decade, evidenced by a skyrocketing expansion of the digital universe as the number of users more than doubled from 2.3 billion in 2012 to 5 billion in 2022, covering 62.5% of the world’s total population. Underlying this impressive figure is an economic and political transformation that saw Big Tech shift from being disrupters to becoming the new capitalist hegemons. Since Facebook’s IPO in 2012, the stock values of Google, Apple, Meta (earlier Facebook), Amazon, and Microsoft — collectively known as GAFAM — increased dramatically. Apple, which led this list of companies in market capitalization, saw its stock value climb sixfold, while Amazon witnessed a twelvefold increase. Between 2017 and 2021, five of the world’s top 10 companies by market cap were the above-mentioned United States-based tech giants. They were followed closely by two Chinese tech companies, Alibaba and Tencent, until 2021. Tesla joined the top 10 in 2020. By 2022, six U.S. tech companies were leading the world’s top-market-cap club compared to only Microsoft in 2000.
These churns in favor of Big Tech, which started well before the pandemic, were further reinforced by the social distancing measures put in place during the health crisis. These measures radicalized digital penetration by dramatically deepening people’s online engagement across the world. Some tech companies, such as Zoom, came into prominence during this very unusual period as remote working (and living) took hold. Still, by any measure, GAFAM was the biggest winner. In 2021, their aggregate profits doubled to USD 320 billion compared with 2019, while sales topped USD 1.4 trillion.
Against the background of this transformation is another, truly anthropological shift, that is currently underway. Take the example of retail trade where this shift is being led by Amazon. The revolutionary promises of digital economic restructuring in the 1990s materialized in the 2010s with the development of e-commerce. According to data from the U.S. Census Bureau, online sales as a percentage of total retail sales grew to 4.2% in 2010 from 0.8% in 2000, and accelerated to 11.8% in the first quarter of 2020. More broadly, e-commerce has steadily increased in the past decade, and has ramped up especially during pandemic lockdowns, attaining new highs in the second half of 2020 in countries where it was already widespread, such as China (25% of retail sales), Germany (21%), and the U.S. (16%). Retail e-sales also accelerated markedly in countries where e-commerce was less developed prior to the pandemic, as well as among new segments of the population. Overall, nearly half of the world’s population is estimated to have made online orders in 2020.
Beyond retail trade, the digitalization of the world under the auspices of Big Tech is an all-encompassing transformation. In 2019, reporter Kashmir Hill tried to live without accessing the services provided by these companies. It was an impossible feat, she concluded, as Big Tech’s reach goes well beyond the products and services branded with the names of tech giants. These companies control internet infrastructure, information flows, and large segments of knowledge management, thereby establishing themselves as de facto (data-driven) intellectual monopolies. Through her experiment, Hill touched directly on the existentialist face of intellectual monopoly, which — as we explain in this essay — is a fundamental economic phenomenon that is only superficially addressed by current policymaking.
The following sections of this essay explore recent attempts to regulate Big Tech as well as the limitations of these regulations, and explain the mechanisms that have led to and reinforced intellectual monopolization. Subsequently, we focus on the emergence of tech giants as data-driven intellectual monopolies while drawing parallels with leading corporations from other industries that hold similar monopolies. Based on this discussion, the final section of the essay suggests an alternative systemic regulatory framework that can more effectively address the intellectual monopoly power of Big Tech.
2. Recent Attempts to Regulate Big Tech
Policy debate on digital capitalism, including the special attention directed towards Big Tech, has largely focused on antitrust regulations and data governance. More recently, governments have imposed digital taxes or a possible minimum tax rate on leading corporations, including tech giants.
In 2019, the US Congress opened an antitrust investigation against Google, Amazon, Facebook, and Apple. The US Congress investigation concluded that these four giants had misappropriated third-party data and indulged in anti-competitive practices. Since then, several prosecutions are underway. For instance, California general attorneys have joined the Federal Trade Commission (FTC) to investigate Amazon for potential abuse of market power and an FTC antitrust lawsuit against Facebook has resumed in early 2022 after an earlier failed attempt.
Moreover, since the Covid-19 pandemic, European countries have become increasingly concerned about potential antitrust violations by U.S. tech giants. In 2021, antitrust regulators in the United Kingdom, Germany, and Australia undertook a joint investigation on the dominance of internet giants. Germany had already blocked Facebook from merging data from its own services in 2019. Among the regulations discussed at the level of the European Union (EU) is the European Commission’s (EC) Digital Markets Act (DMA), which aims to regulate platforms that qualify as gatekeepers because they have what the EC defines as a strong economic, intermediary, and market position in multiple EU countries. Under the DMA, gatekeepers will be fined up to 10% of their total worldwide annual turnover or will be liable to pay periodic penalties of up to 5% of their average daily turnover if they are found engaging in unfair practices towards businesses and customers using their platforms. One of the initial proposals was to break up any technology company that is fined three times over a period of five years.
Towards the end of 2020, China introduced antitrust regulations for digital companies, forcing Ant Group, the financial arm of Alibaba, to delay its Shanghai IPO. This was followed by orders to terminate all financial activities of Ant Group, with the exception of its mobile payments business, a market that in China is dominated by Alibaba and Tencent (94%). Furthermore, the People’s Bank of China accelerated measures to increase the uptake of its digital currency channeled through state-owned commercial banks that would compete against Chinese tech giants’ e-payment platforms AliPay and WeChat Pay. By late 2021, the social distress triggered by the pandemic prompted Chinese President Xi Jinping to call for common prosperity, implemented through a stronger regulation of high incomes, among other measures. Alibaba, Tencent, and other tech giants complied by pledging donations. Thereafter, in the beginning of 2022, China’s State Administration for Market Regulation (SAMR) fined Alibaba and Tencent for not reporting at least 43 acquisitions and for anti-competitive practices. However, the SAMR levied a fine of around USD 80 per case, extremely low compared to the annual incomes of these two companies (around 0.0001% of Tencent’s and 0.00007% of Alibaba’s incomes).
However, even if all the potential regulations discussed above are implemented and if tech giants are found guilty of (and fined for) antitrust or gatekeeper behaviors, these would be, at best, false or insufficient solutions, as the following section explains.
3. Antitrust and Data Privatization: Two False Solutions
In 2017, the legal scholar Lina Khan made the case for a drastic revision of antitrust practices inherited from the anarcho-capitalist era of the Chicago School. Her conclusions echoed what other legal scholars had pointed out earlier — that antitrust law had been narrowed down to only consider consumer welfare, disregarding producers and the overall health of the market, and accepting low prices as a sufficient indicator of sound competition. This is why companies like Amazon, Khan explained, have evaded government scrutiny. Somewhat paradoxically though, Khan’s plea was for a return to the traditional antitrust approach for which competition is a virtue in itself:
The undue focus on consumer welfare is misguided. It betrays legislative history, which reveals that Congress passed antitrust laws to promote a host of political economic ends — including our interests as workers, producers, entrepreneurs, and citizens. It also mistakenly supplants a concern about process and structure with a calculation regarding outcome.
Khan’s work has been hugely influential within and beyond academia. Margrethe Vestager, currently serving as the Executive Vice President of the European Commission for a Europe Fit for the Digital Age and the European Commissioner for Competition, referred to this research before Khan herself became Chair of the FTC in the Joe Biden administration. However, while a recognition of the threat posed by the political and economic power of Big Tech is a positive development, Khan and Vestager’s understanding of competition dynamics is ill-suited to address the current challenges related to the intellectual monopoly (see Section 4). By remaining narrowly focused on markets, even as they consider how artificial intelligence (AI) and Big Data can foster anti-competitive behaviors, antitrust offices have overlooked the effects of knowledge and data concentration, and have inadequately addressed the fact that, more often than not, tech giants operate in natural monopoly markets. This is particularly clear when one looks at the fantasy of a return to small(er) competitive tech and the delusion of individual empowerment thanks to data privatization.
3.1. The Fantasy of Small(er) Tech and the Myth of the “Break-up”
Proposals to reverse the acquisitions that have contributed to tech giants’ market concentration or break up these companies are unlikely to materialize. More crucially, these are false solutions.
First, such regulations will not limit these companies’ chances of working together. As Bengt-Åke Lundvall and Cecilia Rikap argue, tech giants already cooperate for technology. For instance, though Alibaba is formally separated from Ant Group and Alibaba Health, all three companies are controlled by the same holding corporation, and hence can share datasets to boost their individual business. In this way, broken-up companies may still work together, sharing databases and research results to retain their privileged market positions.
Second, and more importantly, competition is not the best solution to concentration in the digital sector. Tech giants’ main businesses engender natural monopoly markets. These are markets in which products become cheaper (and better) when there is only one supplier. Every Google search, every Amazon purchase, and every Facebook or YouTube post contributes to improving the algorithms of these companies and, in turn, the services they provide. The more data processed by the same machine learning model, the better it gets. This creates a preference for centralizing provision in one supplier. Furthermore, there are serious doubts as to whether federated solutions that promote a less asymmetric digital economy can overcome network effects and the quasi-infinite content available on platforms such as Google, YouTube, and Facebook. Overall, competition claims are missing the specificities of this sector and, therefore, will likely fail to provide policies or regulations that truly dismantle the power of tech giants.
Competition is not the best solution to concentration in the digital sector. Tech giants’ main businesses engender natural monopoly markets. These are markets in which products become cheaper (and better) when there is only one supplier…Overall, competition claims are missing the specificities of the tech sector and, therefore, will likely fail to provide policies or regulations that truly dismantle the power of tech giants.
3.2. Against Data Privatization
A similar situation arises when examining data governance. Ostensibly, data privacy legislations seek to protect the privacy of individual citizens and limit data harvesting by tech giants. In reality, they do not limit tech giants, but rather app developers and other intermediary players. Tech giants operate at the infrastructural level of digital capitalism and continue to gather data even in Europe whose General Data Protection Regulation (GDPR) framework is held up as an example for countries around the world. Brett Aho and Roberta Duffield argue that while this legislation does limit cases like the Cambridge Analytica scandal, it does not hamper data harvesting by tech giants. Furthermore, there are multiple loopholes in the regulation, including the fact that most U.S. tech giants have their European base in Ireland, whose Data Protection Commission does not comply with the GDPR.
Besides, tech giants are developing new machine-learning approaches that could potentially enable them to bypass data privacy regulations. One such technique is transfer learning, in which algorithms transfer what they have learned from a data source to other related source domains. For instance, Google has used its large language models — which predict words in a text-based on preceding words — to improve the quality of translations for languages with limited amounts of training data. The transfer learning technique has also been utilized to improve classifications in object recognition and text categorization databases using Amazon data. Additionally, tech giants are leading the development of federated learning, a deep learning approach where algorithms learn while processing data that do not need to be hosted together at the same location.1 See, Benaich, N., & Hogarth, I. (2020). State of AI Report. State of AI. https://www.stateof.ai/; and Jacobides, M.G., Brusoni, S., & Candelon, F. (2021). The Evolutionary Dynamics of the Artificial Intelligence Ecosystem. Strategy Science. The principle is that the training is brought to the device and only trained results (digital intelligence) are harvested. There is also the question of what is being promoted through existing data privacy legislations. By assigning individual property rights to data, these acts foster the privatization and assetization of knowledge. Embedded in these legislations is a misconception that data are created by individuals. In reality, data are generally relational, as explained by Klaus Hoeyer in the context of healthcare data. So-called patient data are actually data of relations and processes, including information on the treating physician, the hospital or clinic, the laboratory delivering test results, etc. The same could be said of an e-commerce purchase, that by definition involves the seller, the consumer, and the platform, or even a social network post, where the reactions of other members craft or are part of the production of the post.
Taking these considerations into account, and recognizing that tech giants keep their big datasets secret, an alternative approach would be to follow the advice of several non-profits and advocacy groups that call for socializing data. This entails allowing legitimate users access to raw aggregated and anonymized datasets following an accreditation process. Users could include states, academic institutions, non-governmental organizations (NGOs), and other firms. However, such initiatives must be accompanied by the socialization of AI algorithms and the installation of the necessary public digital infrastructure. Otherwise, given that the most advanced algorithms and digital infrastructures are concentrated in the hands of tech giants, socializing data would favor them more than any other organization. Armed with the most advanced digital technologies, tech giants have the greatest absorptive capacity for processing and learning from socialized data2 On absorptive capacity, see Cohen, W.M., & Levinthal, D.A. (1990). Absorptive Capacity: A New Perspective on Learning and Innovation. Administrative Science Quarterly. 128–52.
. Conceiving tech giants as (data-driven) intellectual monopolies, the next section offers a deeper understanding of the threats that their emergence and dominance pose to society. On this basis, we provide an alternative policy approach that overcomes the limitations highlighted in this section.
4. The Problem with Intellectual Monopoly Capitalism
The term Intellectual Monopoly Capitalism, coined by the Italian economist Ugo Pagano, refers to the expansion and hardening of intellectual property rights (IPRs), and the inclusion of knowledge among the capital assets of firms. Monopolies, in Paul Baran and Paul Sweezy’s monopoly capital theory, are conglomerates that concentrate and centralize (tangible) capital. Unlike these monopolies, whose advantages are related to the size of their tangible capital, intellectual monopolies significantly and systematically monopolize knowledge, which generally, but not always, contributes to market concentration.
Knowledge is cumulative; we produce knowledge on the basis of existing knowledge. Hence, when access to the most advanced knowledge is curtailed, those with access to it will not only be in a privileged position to innovate but will also develop a greater absorptive capacity to keep innovating anew before others adopt the first innovation. The privilege of the innovator — which takes the form of economic rent — will therefore, become a permanent advantage, perpetuating rentiership over time. Intellectual monopoly, thus, refers to how corporations establish and sustain exclusive control and access to knowledge and information.
When access to the most advanced knowledge is curtailed, those with access to it will not only be in a privileged position to innovate but will also develop a greater absorptive capacity to keep innovating anew before others adopt the first innovation. The privilege of the innovator becomes a permanent advantage, perpetuating rentiership over time. Intellectual monopoly, thus, refers to how corporations establish and sustain exclusive control and access to knowledge and information.
4.1. The Rise of Intellectual Property Rights
For Pagano, the dramatic expansion of IPRs “involves the creation of a legal monopoly that can be potentially extended to the entire global economy”. His claim against a strict intellectual property regime echoes the traditional position of economists treating knowledge as a gratuity. Friedrich Hayek, for example, contends:
The growth of knowledge is of such special importance because, while the material resources will always remain scarce and will have to be reserved for limited purposes, the uses of new knowledge (where we do not make them artificially scarce by patents of monopoly) are unrestricted. Knowledge, once achieved, becomes gratuitously available for the benefit of all.
Among other things, the innovation-cum-economic potential of analyzed data is key to explaining tech giants’ quest for owning and controlling new sources of Big Data. Data extracted by Big Tech companies are not only vast but also diverse, which triggers economies of scope. Diversity amplifies these corporations’ capacity to expand their businesses and innovations.
From this traditional point of view, the expansion of knowledge privatization since the 1980s is particularly worrying. The expansion and hardening of IPRs, first in the U.S., and globally since the mid-1990s with the signing of the Trade Related Aspects of Intellectual Property Rights (TRIPS) Agreement, has certainly enabled the global expansion of intellectual monopolies. TRIPS not only contributed to creating an international IPR regime but also shifted how intellectual property was conceived, from a barrier to trade — as it was considered in the General Agreement on Tariffs and Trade — to a tradable commodity. 3 See, Dreyfuss, R., & Frankel, S. (2014). From Incentive to Commodity to Asset: How International Law Is Reconceptualizing Intellectual Property. Michigan Journal of International Law. 36 (4). pp. 557–602; and Orsi, F., & Coriat, B. (2006). The New Role and Status of Intellectual Property Rights in Contemporary Capitalism. Competition & Change. 10 (2). pp. 162–79. For instance, it was found that just 2,000 corporations own 60 percent of the inventions simultaneously patented at the world’s five largest patent offices between 2014 and 2016.4Dernis, H. et al. (2019). World Corporate Top R&D Investors: Shaping the Future of Technologies and AI. Joint Research Centre and OECD.
IPRs are important restrictions, but secrecy even if it is more discrete, is also a prevailing form of knowledge privatization and assetization. Only 15% of AI papers disclose the code involved. Google’s DeepMind, which is crucial for the company’s AI developments, is among those organizations that usually do not disclose code.
IPRs have been promoted as a means to disclose knowledge. In reality, the holder of the IPR is obliged to license, and thus transfer, the protected piece of knowledge only in cases of compulsory license. Furthermore, the licensees only get access to part of the privatized knowledge because IPR descriptions, particularly in computer-related fields, are usually short and of poor quality.5 Bessy, C. (2019). The Transformations of Conventions for Patent Use and the Role of Legal Intermediaries. HAL Open Science; Lemley, M.A., & Feldman, R. (2016). Patent Licensing, Technology Transfer, and Innovation. American Economic Review. 106 (5). pp. 188–92. Moreover, in order to be effectively adopted, protected innovations need to be integrated into other portions of non-codified knowledge like ‘know-how’, such as the know-how to adapt the technology to the specific requirements of the firm. In fact, a close look at royalties received by tech giants makes it clear that they generally prefer to exploit the knowledge internally without licensing it, which points to the existence of other more pervasive forms of intellectual monopolization.
4.2. Endogenous Dynamics of Knowledge Monopolization
Three additional mechanisms escalate global intellectual monopolization. The first is predation in corporate-scientific networks. This is particularly evident in the pharmaceutical industry where companies rely extensively on the work of scholars and use public funding for their research, but monopolize the profits from commercial exploitation.6See, Mazzucato, M. (2015). The Entrepreneurial State: Debunking Public vs. Private Sector Myths. New York and London: Anthem Press; and Rikap, C. (2019). Asymmetric Power of the Core: Technological Cooperation and Technological Competition in the Transnational Innovation Networks of Big Pharma. Review of International Political Economy. 26 (5). pp. 987–1021. https://doi.org/10.1080/09692290.2019.1620309.
A recent example is Remdesivir, used to treat Covid-19. This drug was patented and sold at an exorbitant price by Gilead, even though its development was based entirely on university research and funding was provided by the U.S. National Institutes of Health (NIH). In fact, the NIH is the most frequently declared external funding source in scientific publications by Pfizer, Novartis, and Roche. Similarly, Google, Amazon, and Microsoft co-authored between 78-87% of their scientific publications until 2019 with universities, but only shared ownership of 0.1- 0.3% of their patents with other organizations.
A second, self-reinforcing intellectual monopoly mechanism relates to the harvesting of data where there is more at stake than privacy concerns. As the former Siemens chief executive Joe Kaeser said, manufacturing and engineering data are “the holy grail of innovation” in many industries. Processed data, called “digital intelligence” by the United Nations Conference on Trade and Development (UNCTAD), orients ongoing businesses and opens up new innovation avenues.
Moreover, within machine learning techniques, deep learning algorithms learn and improve themselves as they process more data. As such, they are a continuously self-improving means of production. Cockburn et al., consider this technology a new method of invention that significantly automates discoveries and expands the types of problems that can be addressed through Big Data analysis. It also reshapes how scientific problems from diverse disciplines are framed. Tech giants have mastered this machine learning technology and, as a result, harvest constant flows of original data sources almost exclusively. Eventually, they could expand their intellectual monopoly without limits based on innovations derived from the digital intelligence derived from processing that data. This is why we argue that tech giants are, in effect, data-driven intellectual monopolies.
Among other things, the innovation-cum-economic potential of analyzed data is key to explaining tech giants’ quest for owning and controlling new sources of Big Data. Data extracted by Big Tech companies are not only vast but also, as noted by Jose van Dijck and Thomas Poell, diverse, which triggers economies of scope (meaning, the chances of monetizing data increase when different data sources are cross-referenced). Diversity amplifies these corporations’ capacity to expand their businesses and innovations.
Furthermore, these companies are centralizing data from third parties. In 2015, Amazon, Microsoft, Google, and Alibaba held in their public clouds 4.9% of the data stored worldwide. By 2020, this proportion had already expanded to 22.8%. When clients process their data with the deep learning algorithms that tech giants offer as a service on the cloud, those algorithms keep learning and getting better as they are used, thus self-improving tech giants’ intangible assets and reinforcing their intellectual monopolies. This means that, even without direct employee access to clients’ data, algorithms can learn from third-party data, expanding tech giants’ intellectual monopolies and allowing them to foray into other industries, from healthcare to transport for instance.
Although tech giants lead this intellectual monopolization mechanism, leading corporations in other sectors are also deploying it widely. The aggressive push towards proprietary data analytics capacities by the BlackRock platform Aladdin and the retail chain Walmart are instances of such mechanisms playing out in finance and retail, respectively.
A third phenomenon is related to the expansion of global value chains. The corollary of the unbundling of productive activities allowed by information and communication technologies (ICTs) is a dramatically increased circulation of information and a related sophistication of information systems, which in turn goes hand in hand with a concentration of the capabilities to govern networks. Lead firms’ planning capacities range from defining the dimensions of each production step taking place in subordinate companies to the setting of norms, standards, and behavioral patterns. Moreover, the uneven distribution of uses of intangible assets along different nodes of the chains allows firms specialized in knowledge-intensive segments to capture most of the gains from scale economies.
Apple’s ‘fabless’ (outsourcing of fabrication) model and its masterful control over supply chains is a case in point. The firm abandoned factories in Colorado Springs and Sacramento in 1996 and 2004 respectively, becoming the most renowned “factory less” goods producer in the world. Most of Apple’s manufacturing is undertaken by firms in China and elsewhere in the Global South, while the company itself operates from “a closed ecosystem where it exerts control over nearly every piece of the supply chain, from design to retail store”. Critical in this panopticon surveying a highly dispersed manufacturing process is the monopoly over intellectual capabilities that allows Apple to capture the lion’s share of the value produced in the chain.
5. Systemic Regulation
Such intellectual monopoly mechanisms among contemporary leading corporations explain the emergence of tech giants, and in turn compel us to undertake an in-depth reassessment of traditional regulatory frameworks. Since this new type of corporation is rooted in the specificities of emerging digital technologies, old recipes like those discussed in Section 2 are doomed to fail.
By criticizing the prevailing regulatory frameworks, we do not pretend to neglect that Big Tech companies tend to have high market power. On the contrary, we argue that to correctly address the concentration of power, the various dynamics of intellectual monopoly must be acknowledged. On this basis, one can take stock of some of the most important transformations of current capitalism and identify relevant policy actions. In this respect, considering that the genesis of tech giants’ power is their intellectual monopolies, we argue that policies should focus on the monopolization of data and knowledge, and aim at democratizing their governance.
The accelerating reach of tech giants’ economic and political power rests on dramatic forces that should not be suppressed but dealt with. In other words, if the future belongs to the (in)visible hand of algorithms, the question that policymakers should answer is, who will operate them? Their responsibility is to take steps to prevent a further concentration of economic value in a few corporate nodes, which is a precursor to an increasing hollowing out of individual and collective human agency.
What is at stake is a concentration of the ability to understand, coordinate, and transform social and economic processes. Collective capabilities that are now harnessed for a profit motive should instead be mobilized to achieve shared social, ecological, and psychological development goals. This requires a new generation of resolute, innovative, and coordinated policies, along at least two main dimensions.
First, following the “do no harm” principle, there must be an emphasis on algorithmic accountability. Responsible algorithmic decision-making should be about more than privacy issues and biases leading to discriminatory and inequitable outcomes. Control over algorithms makes it possible to “model, anticipate, and pre-emptively affect possible behaviors” of individuals and organizations alike, thus allowing tech giants to plan and govern spheres of social life. Furthermore, these capabilities are subjected to powerful monopoly forces. Public authorities must therefore prevent corporate uses of Big Data that encourage detrimental behaviors, such as compulsory consumption, carbon-intensive activities, or online bullying. In the labor process, this detrimental usage includes the algorithmic management and the psychological harassment of workers through the deployment of AI. To that end, large-scale algorithmic apparatuses should be subjected to annual mandatory audits followed by the publication of relevant results. Some uses must be explicitly forbidden, for example, automatic decision-making in work relations. Every worker should have the right to human decision-making on matters related to human resources, such as work, pay, and employment conditions.
Secondly, the resolution of crises and the attainment of socially and ecologically desirable goals must not be curtailed by intellectual monopoly. IPRs should be automatically and generously waived where the free circulation of knowledge can contribute to alleviating social, health-related, and ecological hardships. This type of intermediary policy — intermediary in the sense that the international intellectual property regime would still be in place — would particularly favor peripheral countries in the Global South. States from Global South countries, which are net providers of raw data and net importers of digital services, could join forces to increase their bargaining power against tech giants and demand that they transfer the technology necessary to build data centers using state-owned infrastructure. A step further could be to charge them for privately appropriating data and mandate that such data be stored in local data centers, with the caveat that data is relational and may involve actors from different countries.
Furthermore, intellectual monopolies set science and technology agendas, as was demonstrated by the rise of Big Pharma. This results in innovation rates and directions that privilege profit-making over solving social, ecological, and health crises. At odds with these prevailing experiences, inspiration could be driven from the example of anti-flu vaccine networks which, as Giovanni Dosi points out, have been usually developed as open science. Among other takeaways, this experience shows that global institutional efforts are required to set new research agendas supported by public funds. Considering the transnational nature of information dynamics and the current limited strengths and global articulation of social and political movements, as well as the entanglement of academia with private interests, only a United Nations program could provide a platform for such a global effort in the short term. For instance, it could provide the support to create a global, universal, and free search engine that is also free from advertising. But this is not nearly enough.
In the midst of the pandemic, Google made its community mobility reports temporarily available, which helped to assess the impact of mobility restriction on the spread of the disease. It is shocking that such general interest data are not available on a permanent basis. Given that the ability to process knowledge and behavioral data has become a powerful governing tool, algorithms should be open source and data of general interest publicly available in anonymized forms. A deeper public discussion on the type of data to be retrieved in the first place should also be prioritized. It is through such measures that relevant Big Data arrangements can be deployed to serve public policies and prevent the predation of value across the economic landscape. The Chinese state is already moving in the direction of a data commons in the financial sector. As part of the implementation of its social credit system, the Chinese central bank called data collected by internet platforms a “public good” which should be disclosed and regulated more closely. Aversion to the lack of democracy and pervasive state surveillance in China is no excuse for letting crucial resources for the coordination of social life end up as a private monopoly.
Given that the ability to process knowledge and behavioral data has become a powerful governing tool, algorithms should be open source and data of general interest publicly available in anonymized forms. A deeper public discussion on the type of data to be retrieved in the first place should also be prioritized. It is through such measures that relevant Big Data arrangements can be deployed to serve public policies and prevent the predation of value across the economic landscape.
6. Conclusions
This essay has argued that the contemporary debate on dismantling tech giants offer, at best, an inadequate solution to the rising power of Big Tech. Regulations should be designed to encompass not only market outcomes and data harvesting, but also the full digital technology package (Big Data, algorithms, and infrastructure), with the ultimate aim of dismantling the source of the power of tech giants, namely, their data-driven intellectual monopolies.
Cutting-edge AI algorithms, which are kept secret by tech giants, have already been trained and systematically improved thanks to society’s data. Portions of these algorithms were and are being coded by other organizations participating in tech giants’ innovation networks while most of the associated rents are garnered by the latter. Any move towards a data commons structure will only happen against this backdrop. What is necessary, therefore, is the creation of a commons comprising data, algorithms, and digital infrastructure. This option could tackle both surveillance and data-driven intellectual monopolies. It would be a potential avenue for a socialization that empowers public agencies and private socio-economic actors alike.
A shift in this direction would be a U-turn from the proprietary ideology of the previous fin de siècle. But it would also be a fair return to society. As Big Tech data scientists themselves acknowledge, “algorithms aren’t magic; they simply share with you what other people have already discovered”.
