The epidemiologist Marcel Salathé recommends that Switzerland should adopt the "test-isolate-quarantine" method to combat the Covid-19 pandemic.This includes proximity tracing, i.e. using mobile phone data to trace people who have been in contact with an infected person in recent days in order to detect others who have been infected as quickly as possible and reduce the spread of the virus. This approach has obviously led to a rapid containment of the pandemic. in South Korea and is considered a model in many places.
German Health Minister Jens Spahn wants to locate people who have been infected and their contacts with the aid of mobile phone data; Federal Justice Minister Christine Lambrecht rejects this and knows she has a chorus of data protectionists behind her. The epidemiologists want to limit the suffering caused by Covid-19 disease and the economic and social consequences of plant closures, while data protectionists want to protect civil liberties. Legal experts call for a balancing of interests in such situations. Even in times of a pandemic, the protection of privacy is an issue that can attract attention in a media-overloaded society. As privacy is an emotionally highly charged concept, we risk seeing a battle between rational arguments and the emotions of do-gooders. The decision for or against the use of personal data to contain the pandemic is an excellent case study for the application of ethical guidelines.
What data is at stake?
It is often argued that the data provided by the telecommunications companies is far too imprecise to allow proximity tracing and the notification of people who might be infected, that instead a variety of data sources (credit cards, cameras with face recognition, etc.) is required, and that data protection would no longer be controllable.
With regard to the data of the telecommunications companies, it should be said that this has long been used, for example in conjunction with the personal data of the railway companies, to better understand traffic flows and to make efficient use of public transport, but in fact only offers a very rough grid.
As for the other data required for proximity tracing, the comparison with digital services that are already accepted on a voluntary basis worldwide cannot be ignored. Google Maps users, mostly unwittingly, allow very detailed recording of their movement data, which extends to buildings and even individual rooms. In addition, Google usually also has the contact data of the Maps users, so it can determine which friends were in the same place at the same time. Even without the contact data, Google nonetheless has the technical capability to determine which Maps users were in the same place at a certain time. Google provides us with recommendations when searching for restaurants near our present location that match our preferences derived from past behavior. Tinder and similar apps are on an equal footing with Google in terms of location data and personality profiles.
So why does a large part of the population accept that Google, Apple, and many other digital services use their personal data for a small benefit to be gained from personalized advertising and individual purchase recommendations, but on the other hand, make an outcry when such data is to be used to protect health, the economy, and thus society. If you're calling for privacy, check the Google Dashboard to see what Google knows about you. You should also consider who uses these services in your circle of acquaintances and who has explicitly restricted data recording.
Digital services are complex and overwhelm our decision-making ability.
Google Dashboard is extremely user-friendly, so even digital amateurs can find their way around. But how many people check Google's data and personalization settings before using Maps, Search, Gmail or YouTube? Even if a service provider complies with all data protection regulations (e.g. the GDPR), 99% of consumers find such a task too daunting in view of the amount time or knowledge required.
Data protection laws, ethical guidelines such as those of the IEEE’s Ethically Aligned Design or the general terms and conditions of service providers such as Google constantly use a magic formula, i.e. obtaining the user's consent to the use of data for private purposes. However, this standard recommendation is largely worthless, as the complexity overwhelms the uninitiated and thus merely gives the illusion of transparency and control. It is possible that those responsible are simply shirking the decision as to who may use which data for which purpose.
Feelings are stronger than reason.
Our convenience and knowledge prevent us from informing ourselves sufficiently. Instead, we prefer to read or listen to ready-made opinions that we feel are right, that seem to meet our needs, and that we are therefore happy to adopt. Books like "The Age of Surveillance Capitalism" (Shoshana Zuboff) or "Zucked: Waking Up to the Facebook Catastrophe" (Roger McNamee) open our eyes, but at the same time prepare the ground for fears and do not create the basis for a constructive use of data.
In order to clarify rationally which data should be used for proximity tracing, we must examine which human needs might be positively or negatively affected by the use of this data in the interests of striking a balance. First of all, there is the need for security, which requires protection against arbitrariness such as the deprivation of physical freedom as well as an independent legal system. Then there is the fear of loss of prestige, of status in our community. Will we be ostracized if our friends find out about our state of health? Or do we lose prestige if, as has obviously happened in South Korea, it becomes public that we were present at a meeting of a religious sect? Or will it harm a doctor financially if it becomes known that one of his relatives has tested positive for Covid-19? These threats are countered by the hope of detecting possible contagions early on by quickly tracing back contacts and preventing further infections by quarantining. When we look at the horrific images transmitted from Covid-19 hotspots, it becomes understandable that we put many needs aside for the sake of health. In addition, there is the need for financial provision, which is the prerequisite for almost all other needs, i.e. it is just as important, especially when the food supply is in question, as is currently the case in India, for example.
The decision-makers, whether they are ministers of health or justice, epidemiologists or laymen, must carry out this balancing of interests with a far more detailed and fact-based approach than is possible within the framework of this article. In a situation such as the one we are currently experiencing with the Sars-Cov-2 coronavirus, ethics faces a particular challenge, be it in deciding on the use of personal data or in the triage of Covid-19 patients. On the one hand, it is astonishing how little is discussed publicly about this, but on the other hand it is understandable when one considers that these arguments are far too difficult to be pursued seriously without a long-term perspective. However, this is precisely what ethicists, data protectionists and possibly epidemiologists should do when advising politicians as well as contributing to the formation of public opinion. There is little evidence that this is happening.
Emergency solutions are not solutions.
The Pepp-PT (Pan-European Privacy Protecting Proximity Tracing) project is currently developing a template for country-specific apps. In the view of Netzpolitik.org, a platform for digital freedom rights, this is a "data-saving and epidemiologically preferable alternative" to the use of data from the datenkraken. The Austrian Red Cross already offers a country-specific app for free. The app has the advantage that it only records contacts as such when people are just a few metres apart. Singapore uses a similar app (TraceTogether), which automatically and anonymously stores all such contacts on the mobile phone.
However, in contrast to the use of data from Google, this solution requires users to download an app, a digital handshake when meeting other app users, and a medical confirmation in case of infection. My attempt to get people to perform a digital handshake has only caused uncomprehending head-shaking. Google Play shows more than 50,000 downloads and 461 ratings since March 23, 2020. Even if the app were to be installed 100,000 times and then actually used, this would still only be about 1% of all Austrians.
Much more effective and faster, although not quite as accurate, would be proximity tracing based on navigation data from Google and Apple. Governments might even be able to outsource tracing to Google and monitor it closely. Google confirmed to “The Verge” that they had received inquiries of this nature from health authorities but had declined them as it was not possible to identify a contact from Google data with sufficient certainty. My own experiment with friends to compare our respective location histories on Google brought amazingly reliable results.
If politicians were concerned about a negative reaction from voters, they could even use an SMS or a Google and Apple message to obtain the explicit permission of every citizen. But even Google might be deterred by an undesirable response from the public.
The Corona crisis is an impressive example of the tasks that ethics and other disciplines have to perform in order to safeguard and enhance our quality of life, the dangers of emotionalizing such issues, and what would be possible with a balancing of interests based on a very simple quality of life model.