Between surveillance and technological solutionism: A critique of privacy-preserving apps for COVID-19 contact-tracing

Introduction

Almost immediately following the outbreak of the COVID-19 pandemic, contact-tracing smartphone apps began to emerge as a critical element in national-level control and containment efforts across the globe. Co-opting and harnessing the mundane ubiquity of the smartphone as a sensor for citizen movements and interactions, the first COVID-19 apps were launched in China and South Korea in February 2020, a month before the World Health Organization (WHO) officially declared COVID-19 a pandemic. In the first 6 months of the pandemic, more than 30 countries developed and launched official, government-driven contact-tracing apps in an effort to track and control the spread of the disease within national boundaries. These apps were initiated as part of the public health response to COVID-19. In April 2020, Apple and Google, who together control more than 99% of the global market share for mobile operating systems (StatCounter, n.d.), announced a joint initiative for a privacy-preserving, decentralised contact-tracing development framework and set of specifications called the Google–Apple Exposure Notification (GAEN) framework. The GAEN – originally known as the ‘Privacy-Preserving Contact Tracing Project’ – was launched in May 2020, allowing national health authorities to deploy the framework in their official COVID-19 apps. Following earlier programmes, such as Google Flu Trends in 2008, GAEN marked another foray of technology corporations into digital epidemiology and public health at a global scale. These developments raise concerns about state–industrial surveillance, data sovereignty and digital governance in public health.

In this article, we consider contact-tracing apps not only to be technological fixes within a technological-solutionist paradigm, but also a form of surveillance solutionism occasioning and even necessitating privacy solutionism in response. This is pertinent given technology corporations have increasingly espoused their adherence to privacy-by-design principles (Langheinrich, 2001) and are starting to recognise that they can sell privacy-as-a-feature, suggesting that they can provide greater personal or citizen privacy protection than nation-states, at least in contact-tracing endeavours. We raise questions about how contact-tracing apps (and somewhat paradoxically privacy-preserving versions of them) were positioned as solutions to the global pandemic. Technological solutionism and the techno-fix – grounded in the history and ideology of social engineering – inevitably come with their own problems, too. We also take this opportunity to examine how forms of privacy-preserving surveillance solutions (i.e. those conducted via Bluetooth or in a decentralised manner) operate to introduce, normalise and cement new forms of surveillance via responding to the privacy problems of surveillance solutionism. These considerations become more important with the potential for future combination or integration of various forms of health data (i.e. vaccination status) into contact-tracing apps designed to respond to global health crises.

Our argument proceeds across three parts. First, we examine technological solutionism and its problems before turning our attention to contact-tracing applications as an example of surveillance (and privacy) solutionism in the COVID-19 pandemic. We then draw attention to how technology corporations, such as Apple and Google, have assumed a central role in designing and offering surveillance and privacy solutions (in a way considered to be better than those on offer by states) which extends their infrastructural power via ‘sphere transgressions’ into public health and enables influence of state responses and incursions into state sovereignty (Sharon, 2020; Veale, 2020a, 2020b). Sphere transgressions relate to the way in which technology corporations have been able to leverage their digital expertise into new spheres such as medicine and politics (Sharon, 2020). Third, and finally, we show how privacy is now a marketable aspect of surveillance by technology corporations, and conclude by expressing caution about the ongoing and entrenched adoption of new surveillance technology (even ‘privacy-preserving’ surveillance technologies) as a response to social problems.

The technological fix and its problems: technological solutionism in the digital-surveillant present

COVID-19 contact-tracing apps were designed as a technological fix to a complex socio-medical problem. In the mid-1960s, Alvin M. Weinberg – a pioneering nuclear physicist who had, a couple of decades earlier, worked on the Manhattan Project – coined the term ‘technological fix’ to explore the question of whether technology might replace social engineering. Weinberg (1967) describes the goal of social engineering as ‘invent[ing] the social devices – usually legal, but also moral and educational and organizational – that will change each person’s motivation and redirect his [sic] activities along ways that are more acceptable to the society’ (p. 7). He notes that the technologist is ‘appalled’ by the complexity of the task confronting the social engineer in attempting to make even the smallest social change. In contrast to social engineering, Weinberg (1967) maintains, technological engineering is significantly simpler to implement, measure and evaluate, leading him to question whether ‘social problems [might] be circumvented by reducing them to technological problems’ (pp. 7–8). Given that the tension in Weinberg’s thesis is between two forms of engineering – technological and social – it is unsurprising that his examples of successful technological fixes are ones that are focused on social control and seek to reinforce class, gender and racial lines. For instance, Weinberg (1967) references the intra-uterine device (IUD) as ‘possibly the most important new Technological Fix’, not because it gives women greater individual control over their bodies but because it could be used for societal control over women’s bodies to reduce the birth rate in developing countries by ‘reduc[ing] the individual motivation required to induce a social change’ (p. 8). In another example, Weinberg (1967) proposes a speculative technological fix in the service of racial control and social engineering. Responding to the week-long race riots that broke out in Watts, Los Angeles, 2 years previously in 1965 as a result of police brutality and racial discrimination, Weinberg (1967) suggests a ‘fanciful’ technological fix might be to provide ‘air conditioners and free electricity’ to African American families in Watts ‘on the assumption [. . .] that race rioting is correlated with hot, humid weather’ (p. 9).

Although Weinberg (1967) argues and advocates for the technological fix, he, nevertheless, acknowledges that these fixes rarely solve the problems they are designed to and inevitably create or contribute to new ones in their place. Citing, as an example, the unstable ‘peace’ brought about by the H-bomb, Weinberg (1967) writes, ‘technological solutions to social problems tend to be incomplete and metastable, to replace one social problem with another’ (p. 9). As Linda L. Layne (2000) has written, since Weinberg’s introduction of the technological fix in the 1960s, Science and Technology Studies (STS) scholars have proposed and debated distinctions between technological and social fixes, suggesting alternatives such as a hybrid ‘social-technical fix’ and Layne’s (2000) own proposition of a ‘cultural fix’ (p. 493). The very idea of a ‘fix’, however, as Layne (2000) notes, is ‘problematic, premised as it is on a mechanical model of social life, and ignoring the fact that people may not agree on what needs fixing’ (p. 493). Moreover, the dispute over whether a ‘fix’ is technological, social or cultural elides the fact that ‘all interventions are bound to be simultaneously social, cultural, and technical’ (Layne, 2000: 493).

Weinberg’s legacy can today be seen in the technological solutionism promoted and maintained by Silicon Valley technology corporations (Johnston, 2018) and incisively critiqued by Evgeny Morozov in his 2013 book To Save Everything, Click Here. Morozov (2013) maps out Silicon Valley’s pivot towards amelioration – the idea ‘technology can make us better – and technology will make us better’ (p. x) – over the last decade and the risks this poses. Underpinning this drive for amelioration is, according to Morozov (2013), the ideology of technological solutionism, which, he argues, ‘legitimizes and sanctions’ Silicon Valley’s growing new-found aspirations to recast ‘all complex social situations either as neatly defined problems with definite, computable solutions or as transparent and self-evident processes that can be easily optimized – if only the right algorithms are in place’ (p. 5). More recently, Morozov (2020) has renewed his critique of technological solutionism specifically in response to the COVID-19 pandemic, arguing that solutionist responses to the coronavirus have ‘take[n] the surveillance state to the next level’. Various forms of technology, especially new digital surveillance technologies, ¹ were suggested and adopted as ‘solutions’ to the global COVID-19 pandemic, with contact-tracing apps being arguably the most widespread and notable instance.

Both surveillance solutionism and privacy solutionism – as conceived here – operate within the same political economy and promote the same ‘post-ideological’ ideology (Morozov, 2020) as the overarching discourse of technological solutionism. As Morozov (2020) argues, technological solutionism has emerged out of neoliberalism, replacing its ‘proactive’ ideology with a ‘reactive’ one that – in what is a highly political manoeuvre – deploys technology to sidestep political debate or alternatives (para. 8). The aim and result of technological solutionism, according to Morozov (2020), is to ‘shrink public imagination’, its mandate ‘to convince the public that the only legitimate use of digital technologies is to disrupt and revolutionise everything but the central institution of modern life – the market’. For Morozov (2020), this is nowhere more apparent than in solutionist responses to the COVID-19 pandemic, which, he argues, has turned us all into solutionists and will ‘make it more difficult to imagine a world without the tech giants dominating our social and political infrastructure’ (para. 12). Morozov’s observations also speak to disaster capitalism and the reengineering of society where private industry profits (Klein, 2014, 2020).

Indeed, there are many examples of profiteering through new or old technologies presented as solutions to the COVID-19 crisis or the various problems it has created: contact-tracing apps; vaccination passports; COVID status apps; QR code venue check-in; telehealth; remote work, education and videoconferencing; streaming and home entertainment; online shopping and home delivery; and the cloud storage required for all of this. The consequence is more (or accelerated) data collection and profiling, more incursions into privacy, and simultaneous cementing of the infrastructural power of the already most powerful corporate players (Klein, 2020; Veale, 2020a). French et al. (2020), for instance, draw on critiques of disaster and surveillance capitalism to argue that COVID contact-tracing apps might, in fact, be best thought of as a new form of corporate contract tracing due to their entanglements with Silicon Valley and global corporate giants (p. 2; emphasis in original). Highlighting the potential of corporate contact-tracing to ‘de-center the power of public health authorities’, they raise concerns that this may entrench algorithmic inequality and undermine social justice (French et al., 2020: 2, 4–5).

The COVID-19 disaster has presented new opportunities for governments to expand their power via emergency declarations ² and new laws ³ that allow for expanded surveillance capabilities and technologies. This is evident with regard to arguments that contact-tracing and related technologies, such as QR code venue check-in apps, were essential for reopening the economy and addressing issues of unemployment: (see, e.g., Carr, 2020). Such statements were made by political leaders internationally; for example, former Australian Prime Minister Scott Morrison argued, ‘we need that tool so that we can open up the economy. And that’s why it’s so important’ (as cited by Meixner, 2020: para. 28). The overt connection between contact-tracing and the capitalist economy also has the consequence of introducing and rationalising new forms of surveillance and monitoring; a different manifestation of surveillance capitalism as described by Zuboff (2015, 2019).

Contact-tracing apps: surveillance and privacy solutionism

Contact-tracing as a form of data-driven epidemiology has a long history, and one that cannot be fully traced in this article (see, for example, Bland, 2020; Gostin and Hodge, 1998; Landau, 2021). One critical precursor within this history, however, who draws focus to the locative and data-driven approach that underpinned the development of (and debates around) app-based contact-tracing is the 19th-century English physician John Snow. Snow is considered one of the pre-eminent founders of modern epidemiology (see, for example, Fine et al., 2013)—his pioneering map of the London cholera outbreak of 1854 demonstrated through geographic visualisation the connection between contaminated water supply and cholera, and was instrumental in overturning the prevailing miasmatic theory of contagion at the time (Mitchell, 2011). Moreover, through its location-based and data-driven approach, Snow’s map is considered to be an early example of, and precursor to, contemporary computational and app-based contact-tracing (see, for example, Matthews and Woolhouse, 2005; Pietz et al., 2020; Rainwater-Lovett et al., 2016). Given the importance of geo-location to contact-tracing, it is unsurprising that smartphone apps – which have the potential to exploit the built-in global positioning system (GPS) of smartphones thereby rapidly speeding up if not automating manual processes of contact-tracing – almost immediately became a key response to the COVID-19 crisis. February 2020 saw the launch of the Corona 100 m app in South Korea, which harnessed a smartphone’s GPS to alert the user when they came within 100 m of a location where an infected person had been, and an initial pilot of Alibaba’s Health Code in China, a ‘mini-app’ within its Alipay system, also reportedly tracked location (Liang, 2020; Yang et al., 2020).

Privacy concerns and lack of public trust in governments globally were cited as reasons for low uptake of contact-tracing apps, even in relatively high-trust countries like Singapore, where only 25% of the population reportedly downloaded the TraceTogether app – well below the 60% required for efficacy (Morley et al., 2020). With privacy and trust playing a significant role in the adoption or rejection of contact-tracing apps, an analysis of 48 COVID-19 contact-tracing apps found that over a third of the apps did not have a dedicated privacy policy, and, for those that did, the privacy policies were often ‘opaque, incomplete and impenetrable to the average reader’ (Mousavizadeh et al., 2020: para. 1). Other critiques focus on matters of privacy law and community trust (Greenleaf and Kemp, 2020) or informed consent. Michael et al. (2020), for instance, argue that individuals were not presented with a truly free or informed choice about whether (or not) to download contact-tracing apps, ⁴ especially given arguments for their use were advanced alongside statements about the economic implications of lockdowns, the need to reopen society and the easing of restrictions (i.e. reinstating pre-COVID-19 freedoms). Perhaps the most significant considerations relate to whether contact-tracing applications actually assist with the processes of contact-tracing and are effective in reducing the spread of COVID-19. Manual processes of contact-tracing are distinct from the use of digital tools that send automated notifications on the basis of proximity. There is limited empirical evidence supporting the efficacy of automated mobile phone contact-tracing applications in either identifying previously unknown contacts or reducing the transmission of COVID-19, and efficacy is also predicated on having a high population uptake (see, for example, Braithwaite et al., 2020). An independent evaluation of the Australian COVIDSafe app revealed that the app identified only 17 contacts previously unknown to contact tracers, and that there were high transaction costs as tasks associated with the app added to their workloads without any benefits (Abt Associates, 2021).

In April 2020, and in response to global privacy concerns around contact-tracing, Apple and Google, in what, following Bishop and Green (2008), can be considered an example of ‘philanthro-techno-capitalism’, announced they would together design and release a ‘Privacy-Preserving Contact Tracing’ framework ⁵ for developing contact-tracing apps on Apple iOS and Google Android devices. Similar to TraceTogether and COVIDSafe in Singapore and Australia, respectively, this framework harnesses a smartphone’s Bluetooth rather than its GPS capabilities, but differs from the apps adopted in Singapore and Australia in that it combines this with a decentralised reporting protocol to counter privacy risks relating to centralised data storage and reporting. Notably, when they announced their collaboration on the framework, Apple and Google’s joint statement highlighted solutionism at its core: ‘All of us at Apple and Google believe there has never been a more important moment to work together to solve one of the world’s most pressing problems’ (emphasis added). ⁶ Reflecting upon the approach adopted by Apple and Google, Sharon (2020: 1) highlights the irony given that ‘the questionable data practices that these technology corporations are known for – Google admittedly much more than Apple – would not normally inspire trust for such a sensitive technology like digital contract tracing’ (p. 4).

Launched in May 2020, the framework was later renamed Exposure Notification Framework (ENF), and later, the Google–Apple Exposure Notification Framework. Contact-tracing apps using GAEN offer four key features in a way that is said to preserve a user’s privacy: (a) the user has to opt-in and can opt-out at any time; (b) instead of using exact GPS-based location data, GAEN gathers proximity-based data using randomised Bluetooth IDs; (c) decentralised data storage on the device instead of centralised data storage in the cloud means the app checks Bluetooth IDs associated with positive COVID-19 cases against its own list, and; (d) Apple and Google restrict access to the GAEN to public health authorities and scrutinise GAEN-enabled apps against ‘specific criteria around privacy, security, and data use’. Recognition that Bluetooth is a privacy-invasive technology tends to be neglected in descriptions of the GAEN, for example, Google ⁷ explains:

The Exposure Notifications System does not collect or use the location from your device. It uses Bluetooth, which can be used to detect if two devices are near each other – without revealing where the devices are.

However, the privacy implications of relying on Bluetooth technology had already come to the fore in an earlier example that year. In March 2020, Singapore launched its national-scale TraceTogether app, which, to encourage citizen adoption in the face of growing privacy concerns over GPS location-tracking (Ienca and Vayena, 2020; Michael and Abbas, 2020), used Bluetooth for proximity-based tracking. In April 2020, Australia, following Singapore’s lead, launched the Bluetooth/proximity-based COVIDSafe app, stressing in its campaign material that COVIDSafe ‘will not collect your location information’ (Australian Government, 2020). While GPS-based location-tracking is considered to bring heightened privacy concerns relating to contact-tracing (Kahn, 2020), Bluetooth, too, is known to come with its own privacy impacts (Kostakos, 2008), including concerns about location-tracking. Bluetooth itself can, and does, facilitate location-tracking through the use of beacons – small wireless transmitters that communicate via Bluetooth with other Bluetooth-equipped devices like smartphones and from which granular location can be inferred. Certainly, in the absence of national-scale sensor networks of fixed Bluetooth beacons harnessed to do contact-tracing work, Bluetooth-based contact-tracing apps like Singapore’s TraceTogether and Australia’s COVIDSafe, plus also the GAEN-based apps, register proximity between individual smartphones rather than the explicit location of those devices. And yet, it is critical to note that Bluetooth is a key technology within the location data economy via which precise location data is regularly if not constantly harvested from smartphone users (Mitchell et al., 2022). Although Bluetooth may be generally regarded as more privacy-preserving than GPS tracking, it remains a surveillance technology, with privacy-invasive impacts.

State–industrial surveillance and sovereignty

By mid-2021, more than 49 countries ⁸ had released COVID-19 contact-tracing apps using a range of technologies and approaches, which can broadly be divided into GPS (location-tracking) versus Bluetooth (non-location-tracking), ⁹ and centralised versus decentralised data storage. While a mobile phone’s unique IMEI (international mobile equipment identity) number or SIM card had been explored as complementary means for COVID-19 contact-tracing, it was argued that these cannot be relied upon because users can evade detection by switching their phones temporarily to airplane mode or purchasing ‘burners’, that is, cheap, prepaid mobile phones or SIM cards that can be destroyed or discarded when no longer needed. Grantz et al. (2020) point out that:

Mobile phone data cannot distinguish between multiple people using a single phone or SIM card (either of which may be used as a unique identifier), nor does it account for users with multiple phones or SIM cards, limiting the ability to make any inferences about the behavior of individuals from CDR [Call Detail Records] data. (p. 5)

Furthermore, in addition to using cell phone location data, many health authorities aggregate additional data sources obtained from smart city infrastructure such as facial recognition obtained from CCTV camera and drone footage, as well as credit card and ATM data to track and map people and infections (see, for example, Chen et al., 2020; Foth et al., 2021). The contact-tracing data were combined with quarantine data. For example, India mandated the ‘Quarantine Watch’ app in some parts of the country, requiring quarantined individuals to self-report their whereabouts by uploading ‘their GPS tagged selfie once every hour, with a break from 10 pm to 7 am for sleep’ (Datta, 2020: 235).

Countries that adopted GAEN include Canada (COVID Alert), Denmark (Smittestop), Germany (Corona-Warn-App), Ireland (COVID Tracker Ireland), Italy (Immuni), Japan (COCOA), Latvia (Apturi Covid), Poland (ProteGO Safe), Spain (Radar COVID), Switzerland (SwissCovid), Uruguay (Coronavirus – UY) and New Zealand (NZ COVID Tracer). Other countries opted against implementing GAEN and released their own apps, which largely used centralised data storage, including Australia (COVIDSafe), ¹⁰ Austria (STOPP CORONA), Iceland (Rakning C-19), ¹¹ India (Aarogya Setu), Israel (HaMagen), Norway (smitte|stop), Qatar (EHTERAZ) and Russia (Social Monitoring). France (StopCovid) and the United Kingdom (NHSX) also preferred, at least initially, government-controlled centralised apps. We turn to an examination of the situation in France and the United Kingdom (UK) as they raise questions about the role and power of technology corporations vis-à-vis nation-states.

France was initially in negotiations with Apple and Google regarding access to the GAEN feature that enables Bluetooth operations in the background, without adopting the entire GAEN framework (Kelion, 2020a; Leprince-Ringuet, 2020). Essentially, there were issues of functionality and interoperability with the iOS operating system which meant that the contact-tracing app must have been open to record contact data via Bluetooth, and that the user was unable to use other apps or functionalities of the device (i.e. make phone calls or lock it), while the app was running. The Government of France requested that Apple remove the technical barriers that created problems with app functionality, without implementing additional privacy protections such as preventing the ability for the government to (re)identify who has been issued warnings to be tested due to close contact with other app users that tested positive for COVID-19. Apple refused to do so, meaning that the French government had to develop a way to make the app operational on iOS devices without Apple’s support. At the time, France’s Digital Minister Cédric O was quoted in the media emphasising French and European sovereignty: ‘We’re asking Apple to lift the technical hurdle to allow us to develop a sovereign European health solution that will be tied to our health system’ (as cited by Kelion, 2020a – emphasis added) while noting ‘at a time of domination by the GAFA (Google, Apple, Facebook, Amazon) . . . it seemed important to me to insist on the app’s sovereignty’ (as cited by Leprince-Ringuet, 2020 – emphasis added).

The UK, specifically England, also encountered similar tensions with Apple and Google, first in the development of their own centralised app that sought to store anonymous data in a National Health Service (NHS) database with the view that this would offer advantages for tracing efforts and public health responses. The approach adopted by the UK, at least initially, had a greater focus on distance calculations, which would have provided more epidemiological data and facilitated tracing (e.g. people travelling on public transport). However, trials of this application showed that it was only recognising 4% of Apple phones and 75% of Android devices, as the app was not supported by Apple and Google – because of the same issue France encountered (Kelion, 2020b; Sabbagh and Hern, 2020). The UK government spent months and at least £5 million on developing an app that was unable to operate without Apple and Google’s support (Sabbagh and Hern, 2020). At the time the UK Health Secretary, Matt Hancock, commented that ‘Apple software prevents iPhones being used effectively for contact tracing unless you’re using Apple’s own technology’ and that ‘our app won’t work because Apple won’t change that system . . . and their app can’t measure distance well enough to a standard that we are satisfied with’ (as cited by Kelion, 2020b). Yet, given Apple’s refusal to make changes to the iOS operating system, the UK was effectively forced to abandon its original attempts to develop its own app and use the GAEN framework (Kelion, 2021).

Both situations in France and the UK highlight the power of technology corporations in deciding and directing key features of state responses to public health emergencies. There are some arguments, for example, that centralised databases may be more advantageous from an epidemiological perspective in terms of enabling a greater understanding of the transmission of COVID-19, and in turn, offering improved public health outcomes (Ang, 2020). That is, incursions into privacy may be proportionate, necessary, and supported by public health advice (i.e. a proportionality test when assessing conflicting rights such as privacy and right to health). Irrespective of whether this is the case or not, the key question to pose here is (why) should undemocratically elected multinational (although often US-based ¹² ) technology corporations such as Google and Apple have the power to decide what technical or computational abilities state and health authorities, within their own jurisdictions, have or do not have at their disposal. Concerns about Google and Apple’s involvement in contact-tracing endeavours are not limited to those of privacy; for this situation has created further reliance on undemocratic corporate actors to provide public health services. Our analysis corroborates the commentary by Veale (2020a) who points out that these issues go beyond just privacy concerns: while Google and Apple are not sovereign states, they exercise greater power than (some) nation-states in determining pandemic responses. This then prompts us to connect (below) with larger discourses that critically reflect on the implications not just for the sovereignty of a nation-state but also with regards to more recent notions of data sovereignty (Couture and Toupin, 2019; Mann et al., 2020; Veale, 2020b).

While apps developed using GAEN employ Bluetooth and a decentralised approach to data collection and storage to emphasise their care for privacy, we problematise the issue beyond privacy concerns, while emphasising that surveillance conducted via Bluetooth and with decentralised data storage is still privacy-invasive, despite the claims of Google and Apple regarding their ‘privacy-preserving’ approach. What our examination of the decentralised GAEN approach reveals is the irony inherent in pandemic tech utopias, which are feeding the centralised power held by global and supranational technology corporations, and vested in the technological infrastructure they own, operate and control (Veale, 2020a). This situation is certainly not limited to Apple and Google, with other technology corporations also contributing essential infrastructure for (centralised) contact-tracing apps, too, such as the use of Amazon’s Web Services (AWS) for the cloud storage of proximity data from Australia’s COVIDSafe app (Besser and Welch, 2020). In analysing COVID-19 surveillance solutionism, Morozov (2020) points to the ‘extreme dependence of the actually existing democracies on the undemocratic exercise of private power by technology platforms’ (para. 20). As a number of researchers and commentators have noted, technology corporations are using the coronavirus pandemic to expand their infrastructural power and profit (Foer, 2020; Klein, 2020; Lopatto, 2020; Veale, 2020a). In the current global pandemic, this is most evident in the field of health care, and Couldry and Yu (2018) have previously pointed out that ‘sometimes data and the infrastructures installed to support its collection and use generate new forms of value and reorder relationships between the agents involved in the practices of health care’ (p. 4483).

There are significant consequences beyond privacy, especially in terms of the dependency on big tech infrastructure for public goods, and the way in which this enables further, and ongoing, ‘sphere transgressions’ into many other realms of life (Sharon, 2020). This is achieved not only through the provision of technological infrastructures such as the GAEN – designed explicitly to be interoperable with Google and Apple’s market dominant mobile operating systems – but also through industry funding of research (e.g. Abdalla and Abdalla, 2021). Importantly for our arguments here, privacy-preserving offerings that are engineered as a technofix to the privacy problems associated with surveillance solutionist contact-tracing, and that are provided by big tech (i.e. decentralisation and Bluetooth features that work on their operating systems), actually facilitate and contribute to their expansion of power and ongoing dominance and expansions of surveillance in society. Sharon (2020) observes that Google and Apple themselves have promoted ‘the over-simplified equation of decentralized approaches with privacy-preservation, and centralized approaches with government surveillance’ (p. 7) – whereas perhaps the benefits and trade-offs here are not that clear, at least from an epidemiological perspective – and in this way shows how these big players are extending into political spheres by designing, deciding and providing the solutions to the problems of the global pandemic (Sharon, 2020; Veale, 2020a, 2020b).

Apple and Google are able to maintain their hegemonic position because of the reliance of app developers on interoperability with their operating systems and the platform owners’ supreme role in deciding what apps and which features they will enable (or not) on their platforms. Indeed, these operating systems have been rolled out at global scale, and therefore extend well beyond the reach and power of any individual sovereign nation. This makes it much harder for sovereign nations to be able to challenge Apple and Google, particularly when they work in concert (i.e. true monopoly power across both Android and iOS) (Mann et al., 2020). Apple and Google are leveraging their market power over states (in the examples of France and the UK above), and somewhat paradoxically, using arguments about privacy as a mechanism to do so – in the debates on contact-tracing apps commentators remarked how Apple and Google were ‘portrayed as greater champions of privacy than some democratic governments’ (Sharon, 2020: 1).

While arguments advanced by Apple and Google about privacy preservation may be valid and they may genuinely be working to improve the protection of individual rights, the concern is about what happens when Apple and Google’s intentions are not so benign or benevolent: ‘the drama of contact tracing applications has laid bare how much of both extractive and protective infrastructure is reliant on the choices of a small number of gargantuan corporations’ (Veale, 2020b: 39 – emphasis in original). Perhaps, then, this shows a need for states to work together to address the international duopoly of Apple and Google and restrict their ability to leverage and wield their infrastructure over states in ways witnessed in France, the UK, and elsewhere. ¹³ Veale (2020b) reflects that:

oddly, it is notable that there has been little appetite to attempt to rectify this situation with the legal obligations that sovereign states have at their disposal instead reifying the view of tech giants as state-like themselves, diplomatic interlocutors rather than firms operating under national law. (p. 38)

Challenging surveillance and privacy solutionism

The analysis and critical reflection we present here is not to diminish the significance and utility of attempts in computer science and software engineering to preserve privacy and strengthen the security of contact-tracing apps. The system proposed by the Decentralized Privacy-Preserving Proximity Tracing (DP-3T) project:

provides a technological foundation to help slow the spread of SARS-CoV-2 by simplifying and accelerating the process of notifying people who might have been exposed to the virus so that they can take appropriate measures to break its transmission chain. The system aims to minimise privacy and security risks for individuals and communities and guarantee the highest level of data protection. (Troncoso et al., 2020: 2)

This joins other efforts to improve interface design (Li et al., 2021), better understand user acceptance (Utz et al., 2021) and safeguard against possible vulnerabilities and attacks (Casagrande et al., 2021).

We certainly agree that privacy as a feature of technological devices, and efforts in computer science to create such features, are desirable and welcome. However, dependencies on interoperability between devices and operating systems as well as the technical-epidemiological requirement to deploy the system to a critical mass of users – while essential for the system’s efficacy – can undermine market dynamics and have other less desirable consequences. For example, Ioannou and Tussyadiah (2021) discuss how government-sanctioned or mandated deployments of technology to manage the COVID-19 pandemic can inadvertently increase user acceptance of surveillance, which may remain beyond the immediate public health response. The main focus of our argument is thus not about questioning the merits of any technical efforts to create privacy-preserving systems but, rather, reflecting on the social, political and market conditions of their development and deployment. The implications of our analysis also connect to earlier critiques within the field of surveillance studies that show the various ways that privacy operates as an ally of, and not an antidote to, surveillance (Coll, 2014; Stalder, 2002). The decentralised privacy-preserving contact-tracing framework created by Google and Apple, and that harnesses and promotes Bluetooth technology as less privacy-invasive, demonstrates the ways in which privacy can be used as an argument for, and mechanism supporting, expanded surveillance in society.

Alongside this, and feeding into it, is the growing commodification of privacy by platform corporations. In recent years, Apple has been at the forefront of leveraging privacy as a market advantage. In early 2019, Apple launched a privacy-focused ad campaign for its newest iPhone, declaring ‘Privacy. That’s iPhone’ (Gartenberg, 2019). Within 2 months, Google CEO Sundar Pichai took what was considered to be a thinly veiled swipe at Apple and the elitism inherent in the company’s privacy push. Writing in an op-ed for the New York Times, Pichai (2019) maintained that ‘privacy cannot be a luxury good offered only to people who can afford to buy premium products and services’ and that Google believed ‘privacy must be equally available to everyone in the world’ (para. 5). Since then, Apple has only further cemented its aspirations to be recognised as the smartphone developer of choice for the privacy-conscious consumer. According to Leswing (2021), Apple has positioned itself as the leading privacy-conscious big tech company since at least 2014, but he notes that 2021 has seen Apple ‘turning privacy into a business advantage, not just a marketing slogan’, and argues that ‘Apple’s privacy strategy is now part of its products’ (para. 3). Indeed, Apple’s recent iPhone advertising campaign places personal privacy control front and centre, depicting a personal ‘data auction’ that iPhone user ‘Ellie’ shuts down by using the iPhone’s built-in privacy features, such as App Tracking Transparency and Mail Privacy Protection (Apple, 2022).

Beyond the privacy push of major platform corporations like Apple and Google, Coll (2014) reflects on the ways that ‘privacy has become a precondition for a blossoming economy in the context of the information society’ more generally and that privacy can now effectively be ‘regarded as a tool of governance in service of informational capitalism’ (p. 1250). We contend that privacy is not a panacea to surveillance, in the same way that technology is not the solution to the global pandemic. Rather, privacy operates as a crucial component in broader surveillance structures and indeed one that tends to place the responsibility for protecting it squarely on the individual as a neoliberal subject (e.g. End User Licence Agreements and the entire notice and consent model of privacy) vis-a-vis powerful technology corporations that fundamentally rely upon the business model of surveillance capitalism (Zuboff, 2015, 2019). Therefore, attempting to challenge surveillance solely through privacy is ‘potentially detrimental and can paradoxically reinforce it, since privacy and surveillance are not antagonistic; rather they seem to work together in the deployment of the surveillance society’ (Coll, 2014: 1260). As we have demonstrated, privacy is a marketable aspect of surveillance by technology corporations, and the most effective mechanism in the development and deployment of ‘privacy preserving’ contact-tracing applications by Apple and Google. This not only expands surveillance as a techno-solutionist response to the global pandemic, but also cements these corporations’ infrastructural power in new spheres of activity, such as public health in ways that both exceed the power and infringe upon the sovereignty of nation-states and their democratically elected governments (Sharon, 2020; Veale, 2020a, 2020b).

Conclusion

Contact-tracing apps are but one example of early – and ongoing – technological-surveillant responses to the COVID-19 pandemic, which also include facial recognition technology, thermal cameras, biometric wearables, smart helmets, drones and predictive analytics (Kitchin, 2020). In March 2020, The Economist (2020) published an editorial describing how these technologies combined were creating a ‘coronopticon’ through quarantine enforcement, contact-tracing, flow modelling and social-graph making. Although, at the time of publication, COVID-19 contact-tracing apps’ moment in the spotlight has, for most countries in the world, largely ended, they remain a critical facet of the coronopticon, and a technofix that may be revisited in future pandemic scenarios. Morozov (2020) cautions that crises like the COVID-19 pandemic ‘entrench the solutionist toolkit as the default option for addressing all other existential problems’ (para. 11) as it is easier than addressing root causes, structural or political issues. Similarly, Kim et al. (2021) argue that while the use of contact-tracing apps may disappear (and in many countries have disappeared) post-pandemic, ‘the inclination towards such tech-driven “fixes” in solving global emergencies would endure and revive’ (p. 16). It is for this reason that the COVID-19 crisis should be conceived as not only a public health crisis but also a technocultural one where existing capitalist private–public power relationships are reinforced, and the surveillance society cemented with significant implications: ‘the solutionist responses to this disaster will only hasten the diminishment of our public imagination – and make it more difficult to imagine a world without tech giants dominating our social and political infrastructure’ (Morozov, 2020). All of this feeds into a wider architecture of surveillance that involves ongoing expansion of the powers of both corporations and states to intervene in the private lives of citizens (Yu, 2020). These powers are often not rolled back but continually expanded, through as Morozov would argue, entrenched solutionism with technology and surveillance as the response to social problems. Identifying desirable alternatives requires us to challenge existing technocapitalist landscapes through scholarship that can protect civic sovereignties and democratic systems of polities – and that not only conceive technology or variations of it (i.e. Bluetooth/decentralised) as a solution or ‘quick fix’ but part of the problem.